|This is the talk page for discussing improvements to the Pie chart article.
This is not a forum for general discussion of the article's subject.
|WikiProject Statistics||(Rated Start-class, Mid-importance)|
- 1 Florence Nightingale
- 2 Shouldn't there be a note about the scientific aversion to pie charts?
- 3 scaling of different pie charts?
- 4 these pictures are not polar area diagrams
- 5 Image to Commons?
- 6 Exploding pie chart
- 7 Links
- 8 3D Pie Chart Graphic
- 9 reference  provides no evidence that coxcombs / 'coxcombes' was the name of the book(let)
- 10 Original research? Steven's power law
- 11 Introduction is overly narrow & geeky
- 12 Reader feedback: Upload some useful example h...
- 13 Any evidence for use of the terms "3d pie cake / Perspective pie cake" come from"?
I believe that Florence Nightingale invented this type of chart for her work in epidemiology, can someone confirm this and add the appropriate info? Thanks. --Jimaginator 13:15, Feb 25, 2005 (UTC)
- Stephen Fry said so on a recent broadcast of QI - I was just coming to add it and you had done so already! -- ALoan (Talk) 10:58, 20 November 2005 (UTC)
Shouldn't there be a note about the scientific aversion to pie charts?
I came here searching for a discussion about of how Pie Charts should be avoided because they really skew data. I remember my statistics professor going off on how they should never be used if you want to be taken seriously - but do not remember the reasons why. The external link was helpful - but I would have expected more information on this page. Unfortunately, I'm not really qualified to give such information - so I difer to others.
--Ryan Gardner 22:29, 25 August 2006 (UTC)
- I'm not qualified eive read about this recently – I just cannot remember where. The problem is that people are bad at judging relative proportions when looking at pie charts, making bar charts more suitable when the intention is to indicate absolute amounts rather than general trends. — Lee J Haywood 07:31, 26 August 2006 (UTC)
- Ah, there's an external link at the end of the article that leads to a warning – that's the one I read. — Lee J Haywood 08:43, 26 August 2006 (UTC)
- I will add this to my to-do list. Yes, this should be avoided. Comparisons between graphs are a mess. --Chrispounds 03:53, 28 September 2006 (UTC)
- done. your comments are welcome. --Chrispounds 00:01, 29 October 2006 (UTC)
- The image looks good and I don't see any mistakes in the text. Thanks. — Lee J Haywood 08:20, 29 October 2006 (UTC)
- The comparison illustration makes it quite clear. Thanks for adding this. --Ryan Gardner 23:23, 7 December 2006 (UTC)
- I have just added a reference to Stephen Few essay against the use of pie charts. His work could be used to further expand the article, explaining the cons and pros (there are a few) of pie charts. He also makes a very good case against 3D pies. uiteoi (talk) 13:18, 2 February 2010 (UTC)
- Thanks for the interesting Few reference, it does state the case more clearly than some other explanations. Here, then, is the problem that scientists have with pie charts: that are poor at representing small quantities in relation to the whole, and poor in contrasting similar quantities. But as Few points out, they are such a "natural" for representing simple relationships that they are habitually used to teach young schoolchildren math.
- The questions that arise using a pie chart, then, are: What is the data? Who is the audience? and What message needs to be imparted? It's rather disingenuous to say that a "scientist" can understand a bar chart, but not the raw figures. Surely that's within their profession competence. Almost as surely, they can disambiguate subtle percentages presented in a pie graphic. But not so other readers, who do not deal habitually with figures, who are not accustomed to scrutinizing anomalies in the way that data is collected and represented.
- The scientific argument is: In a perfect world, everyone would be as facile as scientists are at evaluating data, and scientists find bar charts easier to use. The pragmatics are: Only a small part of the population are scientists, and bar charts are less assessable to them. For scientists, they are a minor inconvenience, to general public, pie charts may be the only possible way to appreciate the data. Regards, Piano non troppo (talk) 14:02, 2 February 2010 (UTC)
- As cited by Stephen Few in this essay, several studies have shown, that the general public, not just scientists, can interpret more accurately data presented in Bar Charts rather than Pie Charts, even-though the later is usually perceived as more attractive. In a few cases Pie Charts are appropriate as mentioned by Few, these cases could be emphasized in the article to help readers understand when Pie Charts are appropriate and when they're not. The following quotes could be researched to be used in the body of this article in a "Criticism section":
- "No matter how clever the choice of the information, and no matter how technologically impressive the encoding, a visualization fails if the decoding fails. Some display methods lead to effi cient, accurate decoding, and others lead to inefficient, inaccurate decoding. It is only through scientific study of visual perception that informed judgments can be made about display methods. (William S. Cleveland, The Elements of Graphing Data, Hobart Press, 1994, p. 1)". Cleveland refers to pie charts as “pop charts”.
- "...the perceived area is usually equal to the actual area raised to an exponent of about 0.8, times a scaling constant…In contrast, relative line length [such as the lengths of bars] is perceived almost perfectly, provided that the lines are oriented the same way. (Kosslyn, Stephen, Graph Design for the Eye and Mind, Oxford University Press, 2006, p. 40)"
- "We make angle judgments when we read a pie chart, but we don’t judge angles very well. These judgments are biased; we underestimate acute angles (angles less than 90°) and overestimate obtuse angles (angles greater than 90°). Also, angles with horizontal bisectors (when the line dividing the angle in two is horizontal) appear larger than angles with vertical bisectors. (Naomi Robbins, Creating More Effective Graphs, Wiley, 2005, p. 49)"
- “... 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 pies” (Edward Tufte, The Visual Display of Quantitative Information, Graphics Press, 1983, p. 178.)
- Likewise one could write a section teaching when Pie Charts are appropriate, and I have found at least two cases: i) when the goal is to compare a part to the whole, and ii) when comparing the sum of adjacent parts to another sum of other adjacent parts. The same arguments are also valid for Stacked Area Charts and Stacked Bar Charts.
- I don't have the book right now, but in Edward Tufte's The Visual Display of Quantitative Information he makes solid arguments against Pie Chart's citing French Cartographer that they should never be used. Going to get the book and add the proper citations — Preceding unsigned comment added by PuercoPop (talk • contribs) 02:39, 20 March 2011 (UTC)
scaling of different pie charts?
I removed the following from the article because it isn't explained well enough to understand. If someone knows why different pie charts should be scaled to each other feel free to fix and add it back. It was tacked onto the end of the example section.
Two (or more) pie charts comparing similar data can be created by ensuring that the are drawn to the same scale, using the following formula:
- (radius of b)² / number of data in b = (radius of a)² / number of data in a
-Crunchy Numbers 23:04, 18 September 2006 (UTC)
these pictures are not polar area diagrams
for an example see some of the links in the story or http://flickr.com/photos/ffranchi/286785424/in/pool-16135094@N00/ 188.8.131.52 15:43, 27 November 2006 (UTC)
Image to Commons?
- It is a public domain image so you can move it to commons. However, it would be better to redraw it first, preferably in SVG and with less blank space (e.g. by aligning both graphs horizontally or vertically). It is on my todo list; if you can wait a little bit, I'll try redo it and will upload it to commons afterwards. Schutz 09:20, 22 December 2006 (UTC)
- The data are pretty easy to replicate--just create a bar chart with 22, 21, 20, 19 and 18. Make a replica in a Pie Chart. I did this in PowerPoint and saved it to PNG format for compact display. --Chrispounds 13:46, 24 December 2006 (UTC)
- Schutz's version is a nice improvement. --Chrispounds 14:36, 22 January 2007 (UTC)
Exploding pie chart
I think there ought to be some note about why someone might explode a pie chart. The ressoning seems foreign to me. --Brandon Dilbeck 20:46, 20 January 2007 (UTC)
- 1) To highlight a particular subject, or a group of subjects.
- 2) To clarify a chart with many small pieces.
- 3) For cosmetic purposes. Piano non troppo (talk) 12:53, 30 September 2009 (UTC)
Someone added a commercial link to some pie charts. They are pristine examples of what not to do that I wonder if others want to keep the link just to see what we are avoiding. --Chrispounds 16:16, 22 January 2007 (UTC)
3D Pie Chart Graphic
reference  provides no evidence that coxcombs / 'coxcombes' was the name of the book(let)
the cited page does not provide evidence for the claim that coxcomb diagrams were a misnomer, and it was the books in which the diagrams were published that were called coxcombs. from the page: "These diagrams became known as Florence’s ‘coxcombes’ and were the first pie charts." I did find a source on the web which corroborated the claim: http://www.florence-nightingale-avenging-angel.co.uk/GraphicsPaper/Graphics.htm 184.108.40.206 (talk) 10:35, 23 November 2010 (UTC)
Original research? Steven's power law
The comment about Steven's power law seems like original research: There's no citation.
I also dispute its theoretical grounding: if all area is perceived with a factor of 0.7, this would mean we perceive the total area as 0.7, and each of the sections are perceived as 0.7 of their original. Wouldn't all this mean that the proportions remain the same, meaning the influence of Steven's power law was irrelevant?--220.127.116.11 (talk) 23:02, 2 January 2012 (UTC)
EDIT: I just noticed the previous comment received no replies in over a year, so I took the initiative and deleted the offending comment from the article. I append it here:
- Stevens' power law states that visual area is perceived with a power of 0.7, compared to a power of 1.0 for length. This suggests that length is a better scale to use, since perceived differences would be linearly related to actual differences.
Introduction is overly narrow & geeky
The introduction speaks of amazing geekery, such as: "When angles are measured with 1 turn as unit then a number of percent is identified with the same number of centiturns." Huh? Quite apart from the fact that the word "centiturns" shouldn't be cropping up in an introductory paragraph on anything, saying also that "sectors create a full disk" and that the shape resembles a pie that's been sliced would seem to exclude the inclusion of ring charts and other pie chart variations. I'd like to make the intro more descriptive of the actual content of the article, as per WP guidelines, removing some of fog in the process. --gilgongo (talk) 00:11, 31 December 2012 (UTC)
Reader feedback: Upload some useful example h...
Upload some useful example how we use it in statistical analysis
- Isn't the article chock full of such examples? Seems a totally redundant comment to me. --gilgongo (talk) 15:06, 17 November 2013 (UTC)
Any evidence for use of the terms "3d pie cake / Perspective pie cake" come from"?
The article suggests that the terms for a 3D pie chart, which shows a (superfluous/confusing) 3rd dimension, are "3d pie cake" and "perspective pie cake", but I literally cannot find a single relevant example of these terms anywhere else on the internet, except on pages that are directly quoting this wikipedia page. Even the references at the end of the "3d pie cake / Perspective pie cake" section don't contain those terms anywhere (I used google books search for "cake" in the Good and Hardin book). Did the author of this section just make up terms him/herself??? Does anyone know a standard term for this kind of chart? Just "3D pie chart"? — Preceding unsigned comment added by 18.104.22.168 (talk) 14:27, 6 February 2017 (UTC)