![]() But most of the things that people want to graph are either change in one quantity over time or comparisons between related numbers, with or without time elements. Predictably, it depends on what you need to show. If you must use a pie chart, limit yourself to two or three slices at most, use it only for mutually-exclusive, non-overlapping data, and ensure each “slice” has an immediately apparent relationship against the other(s). Anything more complicated, and the chart falls apart. Slices showing anywhere from 1% to 5% of the circle are especially difficult to judge, and it only gets harder as their number of slices increases. Choose a data viz method that’s a good fit for the relationships you want to discuss.Įffectively use pie charts when showing two mutually-exclusive data sets, like answers to a yes-or-no question. As a result, pie charts can only effectively show the relationships between a few things at once and lack the resolution to display differing magnitudes’ values – a requirement for most natural data sets. But what about a slice that’s one-twentieth the size of the circle? What about a hundredth? Our ability to effectively interact with the circular area breaks down quickly. It’s easy enough to slice a circle in half, after all. You can likely judge what half a circle looks like. Slowing down your audience’s understanding is rarely in your best interest. Don’t make your audience do unnecessary visual work to understand your data. Correctly comparing circular segments is not so easy, requiring more time to make an accurate judgment and arrive at the correct conclusions about the data. A bar chart makes this easy: simply track your eyes horizontally across the page from one bar to another. To compare the relative sizes of pie chart slices, the viewer must mentally superimpose a slice on top of the other. If humans were good at estimating the slice area, we wouldn’t need to do that. Asking the audience for active analysis is asking too much. ![]() Each slice is almost certainly labeled with a percentage. You can prove that people are bad at this on your own: look at a pie chart. ![]() Unfortunately, estimating the circle sector area is the fundamental basis of pie charts. That means humans are fundamentally worse at understanding exponential relationships than linear relationships. But non-linear growth is not nearly as easy for the brain to intuitively grasp. It’s relatively easy to imagine how much longer one foot is than two feet. The human brain is very good at estimating linear relationships. To figure out what makes a good data visualization, we will look at a bad one: what, exactly, makes pie charts so bad? Judgment Calls: But despite our near-universal fondness for the pie chart, they are concealing a dark and fatal flaw: they’re terrible at conveying information. You’ve seen them in newspapers, textbooks, financial reports, and executive summaries. After the bar graph, pie charts are likely the best-known type of data viz. Pie charts are well-loved by visual designers, from corporate boardrooms to telethons. How hard can it be? Consider our old friend the pie chart. Don’t make these common data visualization mistakes. No one will tell you that your simple charts made them doubt your credibility, but they might be thinking about it. At worse, you might find yourself losing deals without knowing why. Murky or confusing data visualization will, at best, reduce your presentation’s effectiveness. ![]() ![]() To make your point effectively, you need reliable charts that instantly reinforce and amplify your messaging. Data visualization is more than just a button in Excel. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |