<?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: Part-to-Whole Help</title>
	<atom:link href="/2014/04/14/part-to-whole-help/feed/" rel="self" type="application/rss+xml" />
	<link>http://helpmeviz.com/2014/04/14/part-to-whole-help/</link>
	<description>Helping people with everyday data visualizations</description>
	<lastBuildDate>Wed, 03 Dec 2014 04:43:09 +0000</lastBuildDate>
		<sy:updatePeriod>hourly</sy:updatePeriod>
		<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=4.0.1</generator>
	<item>
		<title>By: Andrej Lapajne</title>
		<link>http://helpmeviz.com/2014/04/14/part-to-whole-help/#comment-6475</link>
		<dc:creator><![CDATA[Andrej Lapajne]]></dc:creator>
		<pubDate>Thu, 10 Jul 2014 23:37:48 +0000</pubDate>
		<guid isPermaLink="false">http://helpmeviz.com/?p=399#comment-6475</guid>
		<description><![CDATA[Solution C:

IMHO vertical lines - not a good idea. Use bar chart for category comparisons. Also, never use &quot;smoothed&quot; lines in any case, they just create a fiction of some data values that are simply not there.

I would skip all columns except December. The last point in time series is December, so placing the December column immediately to the right adds the structure comparison of products for the last (in this case most important) data point.

See example below. It is very old, but probably catches your idea. Full example is here:
http://www.istudio.si/en/knowledge/gallery/ (see last picture in the DASHBOARDS section)

This example is outdated (7 years old). Today I would use:
1. rolling period instead of year (e.g. dec12..dec13), so that first and last point are comparable
2. enhance the bar chart by also displaying the variance between dec12 and dec13 (so called &quot;Integrated Variance&quot; chart instead of basic bar chart).]]></description>
		<content:encoded><![CDATA[<p>Solution C:</p>
<p>IMHO vertical lines &#8211; not a good idea. Use bar chart for category comparisons. Also, never use &#8220;smoothed&#8221; lines in any case, they just create a fiction of some data values that are simply not there.</p>
<p>I would skip all columns except December. The last point in time series is December, so placing the December column immediately to the right adds the structure comparison of products for the last (in this case most important) data point.</p>
<p>See example below. It is very old, but probably catches your idea. Full example is here:<br />
<a href="http://www.istudio.si/en/knowledge/gallery/" rel="nofollow">http://www.istudio.si/en/knowledge/gallery/</a> (see last picture in the DASHBOARDS section)</p>
<p>This example is outdated (7 years old). Today I would use:<br />
1. rolling period instead of year (e.g. dec12..dec13), so that first and last point are comparable<br />
2. enhance the bar chart by also displaying the variance between dec12 and dec13 (so called &#8220;Integrated Variance&#8221; chart instead of basic bar chart).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Andrej Lapajne</title>
		<link>http://helpmeviz.com/2014/04/14/part-to-whole-help/#comment-6473</link>
		<dc:creator><![CDATA[Andrej Lapajne]]></dc:creator>
		<pubDate>Thu, 10 Jul 2014 23:05:28 +0000</pubDate>
		<guid isPermaLink="false">http://helpmeviz.com/?p=399#comment-6473</guid>
		<description><![CDATA[Hi Ulrik, how about this option: small multiples of &quot;Hills&amp;Valleys&quot; chart?

-- created in Excel with Zebra BI (see http://zebra.bi)

P.S. Some labels do help. Especially when integrated into charts. So different visualization constructions should also be judged by the ability to display labels in a legible way.]]></description>
		<content:encoded><![CDATA[<p>Hi Ulrik, how about this option: small multiples of &#8220;Hills&amp;Valleys&#8221; chart?</p>
<p>&#8212; created in Excel with Zebra BI (see <a href="http://zebra.bi" rel="nofollow">http://zebra.bi</a>)</p>
<p>P.S. Some labels do help. Especially when integrated into charts. So different visualization constructions should also be judged by the ability to display labels in a legible way.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ulrik Willemoes</title>
		<link>http://helpmeviz.com/2014/04/14/part-to-whole-help/#comment-2933</link>
		<dc:creator><![CDATA[Ulrik Willemoes]]></dc:creator>
		<pubDate>Tue, 22 Apr 2014 08:58:37 +0000</pubDate>
		<guid isPermaLink="false">http://helpmeviz.com/?p=399#comment-2933</guid>
		<description><![CDATA[Good comments and suggestions - thanks guys! 
I think the lesson is, that no matter what way you choose to visualize it, multiple chart types would be needed in order to show the both part-to-whole aspect and the inidvidual trends of the products.

Any thoughts on solution C?:]]></description>
		<content:encoded><![CDATA[<p>Good comments and suggestions &#8211; thanks guys!<br />
I think the lesson is, that no matter what way you choose to visualize it, multiple chart types would be needed in order to show the both part-to-whole aspect and the inidvidual trends of the products.</p>
<p>Any thoughts on solution C?:</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Mynda</title>
		<link>http://helpmeviz.com/2014/04/14/part-to-whole-help/#comment-2847</link>
		<dc:creator><![CDATA[Mynda]]></dc:creator>
		<pubDate>Wed, 16 Apr 2014 09:01:10 +0000</pubDate>
		<guid isPermaLink="false">http://helpmeviz.com/?p=399#comment-2847</guid>
		<description><![CDATA[You could use this stacked and separated column chart. It&#039;s similar to a panel chart but you can switch views and chart types. See instructions and sample file here: 

https://sites.google.com/site/e90e50fx/home/stacked-and-separated-dynamic-excel-charts]]></description>
		<content:encoded><![CDATA[<p>You could use this stacked and separated column chart. It&#8217;s similar to a panel chart but you can switch views and chart types. See instructions and sample file here: </p>
<p><a href="https://sites.google.com/site/e90e50fx/home/stacked-and-separated-dynamic-excel-charts" rel="nofollow">https://sites.google.com/site/e90e50fx/home/stacked-and-separated-dynamic-excel-charts</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Matt Lanagan</title>
		<link>http://helpmeviz.com/2014/04/14/part-to-whole-help/#comment-2827</link>
		<dc:creator><![CDATA[Matt Lanagan]]></dc:creator>
		<pubDate>Tue, 15 Apr 2014 05:06:32 +0000</pubDate>
		<guid isPermaLink="false">http://helpmeviz.com/?p=399#comment-2827</guid>
		<description><![CDATA[I would agree with the line chart, however display as small multiples as proposed by Edward Tufte. This could be done with all the data in each chart and a single series highlighted or as a stacked panel chart (e.g. http://peltiertech.com/Excel/Charts/StackedCharts.html), which is cleaner imo and quite close to what you came up with as Option B.

If you have any interaction available then perhaps using this to move a series of interest to the baseline (e.g. http://www.excelcampus.com/charts/dynamic-stacked-column-bar-chart-find-the-missing-trends/) may be useful.]]></description>
		<content:encoded><![CDATA[<p>I would agree with the line chart, however display as small multiples as proposed by Edward Tufte. This could be done with all the data in each chart and a single series highlighted or as a stacked panel chart (e.g. <a href="http://peltiertech.com/Excel/Charts/StackedCharts.html" rel="nofollow">http://peltiertech.com/Excel/Charts/StackedCharts.html</a>), which is cleaner imo and quite close to what you came up with as Option B.</p>
<p>If you have any interaction available then perhaps using this to move a series of interest to the baseline (e.g. <a href="http://www.excelcampus.com/charts/dynamic-stacked-column-bar-chart-find-the-missing-trends/" rel="nofollow">http://www.excelcampus.com/charts/dynamic-stacked-column-bar-chart-find-the-missing-trends/</a>) may be useful.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Joe Mako</title>
		<link>http://helpmeviz.com/2014/04/14/part-to-whole-help/#comment-2812</link>
		<dc:creator><![CDATA[Joe Mako]]></dc:creator>
		<pubDate>Mon, 14 Apr 2014 20:58:12 +0000</pubDate>
		<guid isPermaLink="false">http://helpmeviz.com/?p=399#comment-2812</guid>
		<description><![CDATA[How about a simple line chart?

This way you can see the trends and make useful comparisons, like seeing the two spikes in Product C (the one in March at the expense of A and E, and the one in Dec drawing mainly from A) , and how the pairing of Product A and E changed from tracking to divergent.

Additional context could be gained by adding another chart plotting the actual values for each Product and the overall total by month.]]></description>
		<content:encoded><![CDATA[<p>How about a simple line chart?</p>
<p>This way you can see the trends and make useful comparisons, like seeing the two spikes in Product C (the one in March at the expense of A and E, and the one in Dec drawing mainly from A) , and how the pairing of Product A and E changed from tracking to divergent.</p>
<p>Additional context could be gained by adding another chart plotting the actual values for each Product and the overall total by month.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ed Fitzgerald</title>
		<link>http://helpmeviz.com/2014/04/14/part-to-whole-help/#comment-2808</link>
		<dc:creator><![CDATA[Ed Fitzgerald]]></dc:creator>
		<pubDate>Mon, 14 Apr 2014 16:47:01 +0000</pubDate>
		<guid isPermaLink="false">http://helpmeviz.com/?p=399#comment-2808</guid>
		<description><![CDATA[The B version I think does this pretty well. Make each product a vertical bar chart by month and stack them vertically like you have them.  Maybe add a common reference line across all of them, at say 25%.]]></description>
		<content:encoded><![CDATA[<p>The B version I think does this pretty well. Make each product a vertical bar chart by month and stack them vertically like you have them.  Maybe add a common reference line across all of them, at say 25%.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
