> I had to google to find some arguments against pie charts
Can you give me any arguments for pie charts? I also disagree with that article that the best alternative to a pie chart is a bar chart.
Whenever your doing any sort of data visualization you should always ask "How does this representation convey information that cannot be represented in a table?"
Can you provide an example where a pie chart would tell you more than could be conveyed in a table?
Tables are incredibly information dense, have an unambiguous ordering and can easily scale to expressing 5-6 dimensions of information in the exact same space that a pie-chart can convey. Tables are limited in that they work best when representing a very small number of observations, but pie-charts and bar-charts share this same problem without any of the benefits of tables.
If a table won't cut it, then you probably have enough data that a scatter plot is your best choice. Scatter plots also can easily handle an extra 2-3 dimensions (color, scale, and shape) of information.
Charts are useful for quickly conveying relationships within data, not for presenting data as raw information. They're useful specifically because they're not extremely information dense. If you're making video media, an article, an infographic, a poster, slides for a presentation, or anything similar, tables usually have far less impact and don't convey relationships and context well. Whether charts are the right tool for the job is a communication question. What are you trying to communicate, why, to whom, and in what context? While it perennially annoys the hell out of many in the engineering mindset, people find strategies that take these distinctions into account valuable, and that's why visual designers exist.
To be clear, I'm specifically talking about pie-charts and not all charts. There are plenty of cases where a chart conveys information appropriate to the task.
The relationships are visually represented in a table, the minimum table being just a vertical list. A table can be more, but it isn't required. Additionally the relationships in a table are more clearly represented in most all cases.
Information density also doesn't imply that you most be conveying a lot of information, it relates to the efficiency of your use of visual space.
There are no instances (aside for the other example of a clock plot) where the relationships in your data (since you don't want to use information) are not more clearly conveyed with a list then a pie chart.
> and that's why visual designers exist.
I've spent quite a bit of time working with and talking to visual designers and have yet to meet one who think pie-charts are an effective tool for visual communication.
> To be clear, I'm specifically talking about pie-charts and not all charts. There are plenty of cases where a chart conveys information appropriate to the task.
Ok, let's start with the most obvious. How about a chart showing how much of a pie has been eaten? There's a bell curve of useful applications for pie charts that moves away from that, and the shoulders might be pretty tight, but it's absurd to say that's it's simply wrong for all applications. Even all moderately common applications.
> the relationships in a table are more clearly represented in most all cases
Nonsense. It might be easier to compare specific values, but people simply needing to parse a column of digits makes it the wrong choice for many applications. If your communication problem involves communicating values, a table is a great way to go. If your communication problem involves quickly communicating relationships between a handful of elements, a table is the least effective way to do that. Not everybody has a use case for all data that is benefited by seeing precise values. You can't pretend that's not true by just ignoring that other use cases exist, or that other audiences aren't like you. Go ask an investor if they'd always prefer tables to box whisker graphs, or a statistician if they'd always prefer tables to data plots. Similarly, pie charts are great for showing roughly how several elements comprise a whole. If you wanted to show that something used about 21% of the budget, something else used about 29% of the budget, and something else used about 50% of the budget, putting that in a pie chart quickly communicate the relationship between those values far better than putting those values in a table, bar chart, or many other visualizations. The audience matters. That wouldn't be useful to accountants, but it might be useful to give a a company-wide audience a rough idea of why some executive action was enacted, for example. The purpose of data visualizations is to present things in ways that reduce the cognitive load of parsing values while still communicating the overall purpose. A box and whisker graph wouldn't be useful for that visualization. And a to the vast majority of people, table would be less efficient at communicating that idea than a pie chart.
> Information density also doesn't imply that you most be conveying a lot of information, it relates to the efficiency of your use of visual space.
Without a specific communication problem to solve, you can't determine what the best use of space is. You're expecting every audience to parse information like you do. You're wrong. That's why visual designers exist. The oft-repeated design maxim of avoiding wasted space is something only non-designers say because it's much easier to gauge and reason about than the most effective use of space. Optimizing for economy of space isn't even useful in many contexts. When the purpose is efficiently communicating something, giving the primary message focus often means giving it a lot of space, be it conceptually or physically.
> There are no instances (aside for the other example of a clock plot) where the relationships in your data (since you don't want to use information) are not more clearly conveyed with a list then a pie chart.
Ok, how about how many slices of pie have been eaten?
> I've spent quite a bit of time working with and talking to visual designers and have yet to meet one who think pie-charts are an effective tool for visual communication.
Frankly, people like you-- really strong opinions on visual design but have never practiced it, don't know the methodologies, don't know what most people's use cases are for data visualizations, and generally don't understand why everybody doesn't like looking at numbers as much as you-- are probably people that most experienced visual designers don't want to talk about charts with. Regardless, you could tell me that everyone you've ever met hated pie charts and that still wouldn't be worth any more in this context than just saying it's your arbitrary preference.
Working as a long-time developer, I did encounter a few tutorial-drunk designers who tried to assert their ideas about proper software development in dev meetings they happened to be at. As a developer, it was always like "ok there fella. We'll take your opinion into account. Now go play in photoshop." It was pretty rare, though. But I'm continually astonished by the number of developers that assume their amazing superpowers of reason and logic make them qualified to tell any credentialed experienced professional how their field works.
I'm not sure I can give you an example where a pie chart gives you more info than a take. What a pie chart does do is convert information fairly quickly.
Personally I think you should use it basically for percentages only (or similar situations) it makes it fairly easy to see if one piece of pie is larger than another. Sure a bar chat y does the same, except a bar chart didn't let you easily know the percentage. Sure you can say 25% in the label or something. But a pie chart makes it obvious.
A table is cool, but it takes longer to process it.
I like them too. Creating one with d3 + svelte is not very difficult, which then allows you to apply your own customization. I've found the demo at the bottom of this page a good place to start.
Even then, it's pretty ridiculous to say that this is a valid argument against using pie charts. It's a valid argument against using them when you've got multiple similarly sized slices, or some other thing that would make them the wrong choice, but there are great reasons to not use most visualizations in incompatible contexts. Just like anything else in graphic design, these things should be used when they are better at conveying a point than other methods and not used when they aren't. Coming up with rules like this is like saying "never use garlic" and then coming up with every dish from chocolate pudding to iced tea that's not made better with garlic. Use the right tool for the job. It's design "rules" like this that will inspire some insecure "Ha! Gotcha! So-called expert!" type of nerd rant from a developer in a meeting where I'm presenting work I've done as a designer.
>Can you give me a single example where a pie chart would better convey information than a table?
there's a specific type of pie chart that's good: a "clock chart".
this is when you're depicting how long some contiguous sequence of events took as a fraction of a whole duration. you map the start and end to the 12 o'clock position, and the intervening ones wrap around clockwise.
e.g. the history of the universe with the big bang at the top, then formation of stars, galaxies, planet earth, dinoaurs etc as you go around clockwise. then human history is a tiny slice at like 30 seconds to midnight or whatever it is.
this works because it piggybacks on people's existing strong spatio-temporal intuition for clock faces. time is a flat circle.
> You're ignoring the reason data visualizations exist in the first place.
My opinion on pie-charts and poor visualization comes from years of studying data visualization, not ignoring it. Would you argue that Tufte doesn't value data visualization since he holds the same views?
Tufte can back up what he says with reason and examples, rather than repeatedly making the same assertion. Just saying "well Tufte says so" is a pretty blatant appeal to authority.
> In the adjacent pie chart, try to figure out which group is the biggest one [...]. You will probably struggle to do so and this is why pie charts must be avoided.
It's not that bad when the purpose of the pie is not to figure out which group is the biggest.
I find the pie pretty appropriate when the goal is to figure out
- if all the groups are more or less the same
- if one or several groups are way bigger than the other groups
- if one or several groups are way smaller than the other groups
Not saying pies are the best for this (bar charts work well too), just saying it works well.
> try to order them by value
I mean, why would you not order them already inside the chart? So that the reader doesn't have to do it.
> It's not that bad when the purpose of the pie is not to figure out which group is the biggest
But it's not better than a sorted table:
- A (90%)
- B (5.1%)
- C (4.9%)
Which group there is biggest? Which is the smallest?
> why would you not order them already inside the chart?
What's the implicit ordering of a pie chart? It's not hard to define one that makes sense (start at midnight) but it's not implicit in the chart, it needs to be explained to the viewer. Since similar sized portions of the chart are hard to distinguish, there's no way to be visually certain that you are correctly understanding the ordering.
There is argument to be made for a kind of carousel: variant selection and product photos in shops.
When I shop for clothes online, I click through colors and sometimes through photos to see clothes from different angles or zoomed in. It's a sneaky kind of carousel that is actually useful.
I went looking for the famous Edward Tufte quote about pie charts, Tufte being one of the elder statesmen of data visualization, and found it at this old HN thread:
A table is nearly always better than a dumb pie chart; the
only worse design than a pie chart is several of them, for
then the viewer is asked to compare quantities located in
spatial disarray both within and between charts [...] Given
their low density and failure to order numbers along a
visual dimension, pie charts should never be used.
In HCI, I think the general (and researched) consensus about pie charts is that they are fine and probably perform better than alternatives when they have a few (as in probably 4 or less) slices, but then go to total crap for anything much more. I think one of the best essays on this is from Stephen Few, but there's a fair amount of research on the Wikipedia page too.
Being against pie charts is a great example of mid-wit (angry, ranting about nothing). The anti pie chart crowd recognises that other plots can achieve everything the pie chard does, and more. But the fail in this understanding is the assumption of "more is more". Sometimes you don't want readers getting down into the details. Sometimes you don't want readers comparing the relative sizes of pie segments (which the human eye tends to do on bar charts). Sometimes pie charts just suit the overall layout of the document they're part of.
tl;dr be mindful there's often a better plot to use than a pie chart, but if you want to use a pie chart, go for it.
I like 2 slice pie charts for data where those 2 pieces encompass the whole. Bar charts are superior in every other way (3+ segments, easier to compare, etc.), but are not as good at conveying you're looking at the total.
https://www.data-to-viz.com/caveat/pie.html