<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:wfw="http://wellformedweb.org/CommentAPI/"
	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/"
	xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
	xmlns:creativeCommons="http://backend.userland.com/creativeCommonsRssModule"
xmlns:media="http://search.yahoo.com/mrss/" 
>

<channel>
	<title>Graphic Sociology &#187; voting</title>
	<atom:link href="http://thesocietypages.org/graphicsociology/tag/voting/feed/" rel="self" type="application/rss+xml" />
	<link>http://thesocietypages.org/graphicsociology</link>
	<description>Analyzing the visual presentation of social data. Each post, Laura Nor&#233;n takes a chart, table, interactive graphic or other display of sociologically relevant data and evaluates the success of the graphic.</description>
	<lastBuildDate>Thu, 12 Feb 2015 19:50:01 +0000</lastBuildDate>
	<language>en-US</language>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	
<copyright>Copyright 2007-2015 Graphic Sociology</copyright>
	<creativeCommons:license>http://creativecommons.org/licenses/by-nc-sa/3.0/</creativeCommons:license>
	<item>
		<title>Poll data, poll dancing</title>
		<link>http://thesocietypages.org/graphicsociology/2010/06/03/poll-data-poll-dancing/</link>
		<comments>http://thesocietypages.org/graphicsociology/2010/06/03/poll-data-poll-dancing/#comments</comments>
		<pubDate>Thu, 03 Jun 2010 11:50:42 +0000</pubDate>
		<dc:creator><![CDATA[Laura Norén]]></dc:creator>
				<category><![CDATA[Uncategorized]]></category>
		<category><![CDATA[Europe]]></category>
		<category><![CDATA[political]]></category>
		<category><![CDATA[voting]]></category>

		<guid isPermaLink="false">http://thesocietypages.org/graphicsociology/?p=848</guid>
		<description><![CDATA[What works In the words of the creator of this graphic, the point here is that &#8220;there is no pattern&#8221;. The YouGov pollsters seemed to be a little more accurate, but then, as was also pointed out by the graphic&#8217;s creator, they only had one year to give it a go. Low N on that [&#8230;]]]></description>
				<content:encoded><![CDATA[<div id="attachment_849" style="width: 610px" class="wp-caption aligncenter"><a href="http://www.flickr.com/photos/25541021@N00/4583071265/sizes/o/"><img src="http://thesocietypages.org/graphicsociology/files/2010/06/poll_dancing.jpg" alt="UK elections poll data | Information is Beautiful" title="UK elections poll data | Information is Beautiful" width="600" height="874" class="size-full wp-image-849" /></a><p class="wp-caption-text">UK elections poll data | Information is Beautiful</p></div>
<h3>What works</h3>
<p>In the words of the creator of this graphic, the point here is that &#8220;there is no pattern&#8221;.  The YouGov pollsters seemed to be a little more accurate, but then, as was also pointed out by the graphic&#8217;s creator, they only had one year to give it a go.  Low N on that group, but maybe we can call them &#8216;one to watch&#8217;.  </p>
<p>There is always a tendency in science &#8211; bench science, social science, any kind of science &#8211; to show positive results.  It sort of sounds like:  &#8220;Look!  I found something!&#8221; Or, more likely, &#8220;After controlling for everything I could think of, including maternal grandmother&#8217;s underwear size, I have found a statistically significant correlation in the predicted direction.&#8221;  But there is almost no support for saying, more or less, &#8220;I was looking for something but I found nothing.&#8221;  In this particular case, a non-finding is of interest because it suggests action. We can stop paying attention to prediction polls (or chance it and continue to pay attention to YouGov, with a grain of salt).  What works best here is the rigorous reporting of no pattern.  Multiple polling companies, multiple elections, still no pattern.  </p>
<h3>What needs work</h3>
<p>Seriously needs a key.  Red and blue are always political colors, yellow not necessarily so, and the meanings of each cannot be assumed.  </p>
<p>Love the title &#8216;poll dancing&#8217; but wish it would mention &#8216;UK&#8217; and &#8216;elections&#8217; somewhere.  We can deduce from the listing of the Guardian as a source that it probably has something to do with the UK, but information is global now, and we cannot assume national origins anymore.  I often make this mistake myself, easy to forget to mention the nation-state.  The good news is that our audiences are no longer only our neighbors.  Or at least that&#8217;s how I like to think of it.</p>
<h3>References</h3>
<p><u>Suggestion from</u><br />
Momin, a young fellow who contacted me by email suggested I post this one.</p>
<p><u>Graphic</u><br />
McCandless, David and Key, James. (2010) <a href="http://www.guardian.co.uk/news/datablog/2010/may/06/general-election-2010-opinion-polls-information-beautiful">&#8220;Poll Dancing:  How accurate are poll predictions?&#8221;</a> from <a href="http://www.informationisbeautiful.net/">Information is Beautiful</a>.</p>
<p>See also:<br />
McCandless, David. (2010, May 6) <a href="http://www.guardian.co.uk/news/datablog/2010/may/06/general-election-2010-opinion-polls-information-beautiful">General election 2010: Information is Beautiful goes poll dancing</a> at <a href="http://guardian.co.uk/">The Guardian</a>, <a href="http://www.guardian.co.uk/news/datablog">Data Blog.</a></p>
<p><u>Data</u><br />
<a href="http://bit.ly/polldancing/">http://bit.ly/polldancing</a></p>
]]></content:encoded>
			<wfw:commentRss>http://thesocietypages.org/graphicsociology/2010/06/03/poll-data-poll-dancing/feed/</wfw:commentRss>
		<slash:comments>3</slash:comments>
		<creativeCommons:license>http://creativecommons.org/licenses/by-nc-sa/3.0/</creativeCommons:license>
<media:content url="http://thesocietypages.org/graphicsociology/files/2010/06/poll_dancing.jpg" width="" height="" medium="image" type="application/octet-stream" />	</item>
	</channel>
</rss>
