<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Twitter [horrific] Graph!</title>
	<atom:link href="http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/feed/" rel="self" type="application/rss+xml" />
	<link>http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/</link>
	<description></description>
	<lastBuildDate>Mon, 08 Mar 2010 16:11:22 -0500</lastBuildDate>
	<generator>http://wordpress.org/?v=2.9.2</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Jon Peltier</title>
		<link>http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/comment-page-1/#comment-2294</link>
		<dc:creator>Jon Peltier</dc:creator>
		<pubDate>Tue, 11 Nov 2008 14:57:14 +0000</pubDate>
		<guid isPermaLink="false">http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/#comment-2294</guid>
		<description>Ron -

&quot;... it’s not intuitive from looking at the bar chart...&quot;

Given the context of the pie chart, and the way a lot of amateur pie charts are constructed, it&#039;s not obvious for the pie chart either. The bar chart gives the option to cleanly show either, but a notation should be required in either case. One assumes that the pie chart disallows multiple responses, but I&#039;ve seen this rule violated.

The data label on the bar tells me the bar reaches to 59%. The chart implies 59%, but if you look at it too quickly, you may infer 100%.</description>
		<content:encoded><![CDATA[<p>Ron -</p>
<p>&#8220;&#8230; it’s not intuitive from looking at the bar chart&#8230;&#8221;</p>
<p>Given the context of the pie chart, and the way a lot of amateur pie charts are constructed, it&#8217;s not obvious for the pie chart either. The bar chart gives the option to cleanly show either, but a notation should be required in either case. One assumes that the pie chart disallows multiple responses, but I&#8217;ve seen this rule violated.</p>
<p>The data label on the bar tells me the bar reaches to 59%. The chart implies 59%, but if you look at it too quickly, you may infer 100%.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ron Shevlin</title>
		<link>http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/comment-page-1/#comment-2286</link>
		<dc:creator>Ron Shevlin</dc:creator>
		<pubDate>Mon, 10 Nov 2008 17:03:18 +0000</pubDate>
		<guid isPermaLink="false">http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/#comment-2286</guid>
		<description>I prefer the pie chart. 

First off, it&#039;s not intuitive from looking at the bar chart whether or not multiple responses were allowed.

Second, the way the bar chart is drawn, the 59% bar stretches across the whole area allotted for the chart. Visually, this implies something closer to 100%, not 59%. That&#039;s never a problem with a pie chart.</description>
		<content:encoded><![CDATA[<p>I prefer the pie chart. </p>
<p>First off, it&#8217;s not intuitive from looking at the bar chart whether or not multiple responses were allowed.</p>
<p>Second, the way the bar chart is drawn, the 59% bar stretches across the whole area allotted for the chart. Visually, this implies something closer to 100%, not 59%. That&#8217;s never a problem with a pie chart.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: DSA</title>
		<link>http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/comment-page-1/#comment-2046</link>
		<dc:creator>DSA</dc:creator>
		<pubDate>Sat, 21 Jun 2008 00:20:12 +0000</pubDate>
		<guid isPermaLink="false">http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/#comment-2046</guid>
		<description>Jon - Good point.  

I would also recommend the two Stephen Few books above.</description>
		<content:encoded><![CDATA[<p>Jon &#8211; Good point.  </p>
<p>I would also recommend the two Stephen Few books above.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jon Peltier</title>
		<link>http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/comment-page-1/#comment-2045</link>
		<dc:creator>Jon Peltier</dc:creator>
		<pubDate>Fri, 20 Jun 2008 22:17:43 +0000</pubDate>
		<guid isPermaLink="false">http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/#comment-2045</guid>
		<description>I would also suggest two books by Stephen Few: &lt;a href=&quot;http://www.amazon.com/gp/redirect.html%3FASIN=0596100167%26tag=peltiertechni-20%26lcode=xm2%26cID=2025%26ccmID=165953%26location=/Information-Dashboard-Design-Effective-Communication/dp/0596100167%253FSubscriptionId=02E5W5871AJF7PMMMS82&quot; rel=&quot;nofollow&quot;&gt;Information Dashboard Design: The Effective Visual Communication of Data&lt;/a&gt; and &lt;a href=&quot;http://www.amazon.com/gp/redirect.html%3FASIN=0970601999%26tag=peltiertechni-20%26lcode=xm2%26cID=2025%26ccmID=165953%26location=/Show-Me-Numbers-Designing-Enlighten/dp/0970601999%253FSubscriptionId=02E5W5871AJF7PMMMS82&quot; rel=&quot;nofollow&quot;&gt;Show Me the Numbers: Designing Tables and Graphs to Enlighten&lt;/a&gt;. Few&#039;s books are more approachable and more practical than Tufte&#039;s.</description>
		<content:encoded><![CDATA[<p>I would also suggest two books by Stephen Few: <a href="http://www.amazon.com/gp/redirect.html%3FASIN=0596100167%26tag=peltiertechni-20%26lcode=xm2%26cID=2025%26ccmID=165953%26location=/Information-Dashboard-Design-Effective-Communication/dp/0596100167%253FSubscriptionId=02E5W5871AJF7PMMMS82" rel="nofollow">Information Dashboard Design: The Effective Visual Communication of Data</a> and <a href="http://www.amazon.com/gp/redirect.html%3FASIN=0970601999%26tag=peltiertechni-20%26lcode=xm2%26cID=2025%26ccmID=165953%26location=/Show-Me-Numbers-Designing-Enlighten/dp/0970601999%253FSubscriptionId=02E5W5871AJF7PMMMS82" rel="nofollow">Show Me the Numbers: Designing Tables and Graphs to Enlighten</a>. Few&#8217;s books are more approachable and more practical than Tufte&#8217;s.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: DSA</title>
		<link>http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/comment-page-1/#comment-2044</link>
		<dc:creator>DSA</dc:creator>
		<pubDate>Fri, 20 Jun 2008 21:08:00 +0000</pubDate>
		<guid isPermaLink="false">http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/#comment-2044</guid>
		<description>Yuvi - I would like to commend you on your positive attitude!  My post was not aimed at criticizing you, but at the chart layout.  There were other charts on your page that were much better.

I may suggest, The Visual Display of Quantitative Information by Edward Tufte if you are serious about improving data visualizations.

Thank you for the comment!</description>
		<content:encoded><![CDATA[<p>Yuvi &#8211; I would like to commend you on your positive attitude!  My post was not aimed at criticizing you, but at the chart layout.  There were other charts on your page that were much better.</p>
<p>I may suggest, The Visual Display of Quantitative Information by Edward Tufte if you are serious about improving data visualizations.</p>
<p>Thank you for the comment!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jon Peltier</title>
		<link>http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/comment-page-1/#comment-2043</link>
		<dc:creator>Jon Peltier</dc:creator>
		<pubDate>Fri, 20 Jun 2008 14:46:05 +0000</pubDate>
		<guid isPermaLink="false">http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/#comment-2043</guid>
		<description>Yuvi -

I like your enthusiasm. I was merely pointing out a few ways to make the charts easier for the reader to interpret. Keep &#039;em coming.</description>
		<content:encoded><![CDATA[<p>Yuvi -</p>
<p>I like your enthusiasm. I was merely pointing out a few ways to make the charts easier for the reader to interpret. Keep &#8216;em coming.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Yuvi</title>
		<link>http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/comment-page-1/#comment-2042</link>
		<dc:creator>Yuvi</dc:creator>
		<pubDate>Fri, 20 Jun 2008 08:13:35 +0000</pubDate>
		<guid isPermaLink="false">http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/#comment-2042</guid>
		<description>Hi folks. I&#039;m the guy behind that &quot;totally horrible and ineffective chart&quot;.

I will learn from my mistakes folks. So, keep pointing them out (Thanks for the comments on the Blog, Jon)</description>
		<content:encoded><![CDATA[<p>Hi folks. I&#8217;m the guy behind that &#8220;totally horrible and ineffective chart&#8221;.</p>
<p>I will learn from my mistakes folks. So, keep pointing them out (Thanks for the comments on the Blog, Jon)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jon Peltier</title>
		<link>http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/comment-page-1/#comment-2026</link>
		<dc:creator>Jon Peltier</dc:creator>
		<pubDate>Mon, 16 Jun 2008 21:54:27 +0000</pubDate>
		<guid isPermaLink="false">http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/#comment-2026</guid>
		<description>The pie with labels instead of a legend was a different data set.

I decided to take it easy on the blog author, since the bio said he&#039;s 17 or so. A few friendly comments, since he&#039;s had no exposure to anything important yet.</description>
		<content:encoded><![CDATA[<p>The pie with labels instead of a legend was a different data set.</p>
<p>I decided to take it easy on the blog author, since the bio said he&#8217;s 17 or so. A few friendly comments, since he&#8217;s had no exposure to anything important yet.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: DSA</title>
		<link>http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/comment-page-1/#comment-2025</link>
		<dc:creator>DSA</dc:creator>
		<pubDate>Mon, 16 Jun 2008 21:22:32 +0000</pubDate>
		<guid isPermaLink="false">http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/#comment-2025</guid>
		<description>@Jon - Sorting the graph in descending order does make for a better visual.  I did look back at their site and they do include another version of the pie with values AND labels.  Maybe a special post is due on &quot;how to make horrific graphs a little less horrific.  :)

@Dan - As Jon pointed out, they do have another version with labels instead of a legend.  I think a bar chart, sorted in descending order, is much more effective than the best of pies.

I think the only time I might (slight possibility) use a pie chart is when I am only comparing two values.  Even still, a column/bar chart would probably be more effective and take up less space.

More than anything, this post is just a reminder of what not to do.  If you choose to use a pie chart, don&#039;t make the same mistake and have a legend like the one above.

Thank you for your comments and feedback!</description>
		<content:encoded><![CDATA[<p>@Jon &#8211; Sorting the graph in descending order does make for a better visual.  I did look back at their site and they do include another version of the pie with values AND labels.  Maybe a special post is due on &#8220;how to make horrific graphs a little less horrific.  <img src='http://supportanalytics.com/blog/wp-includes/images/smilies/icon_smile.gif' alt=':)' class='wp-smiley' /> </p>
<p>@Dan &#8211; As Jon pointed out, they do have another version with labels instead of a legend.  I think a bar chart, sorted in descending order, is much more effective than the best of pies.</p>
<p>I think the only time I might (slight possibility) use a pie chart is when I am only comparing two values.  Even still, a column/bar chart would probably be more effective and take up less space.</p>
<p>More than anything, this post is just a reminder of what not to do.  If you choose to use a pie chart, don&#8217;t make the same mistake and have a legend like the one above.</p>
<p>Thank you for your comments and feedback!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Dan</title>
		<link>http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/comment-page-1/#comment-2024</link>
		<dc:creator>Dan</dc:creator>
		<pubDate>Mon, 16 Jun 2008 15:22:59 +0000</pubDate>
		<guid isPermaLink="false">http://supportanalytics.com/blog/2008/06/twitter-horrific-graph/#comment-2024</guid>
		<description>The pie would work reasonably if they&#039;d just taken the time to label the segments, rather than using a legend.

Aggregating the smaller catergories and just showing the top 6 (&gt;=5%) and &quot;Other&quot; would also have made it easier.</description>
		<content:encoded><![CDATA[<p>The pie would work reasonably if they&#8217;d just taken the time to label the segments, rather than using a legend.</p>
<p>Aggregating the smaller catergories and just showing the top 6 (&gt;=5%) and &#8220;Other&#8221; would also have made it easier.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
