<?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/"
	>

<channel>
	<title>Compare Stuff News &#187; Validation</title>
	<atom:link href="http://blog.compare-stuff.com/category/validation/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.compare-stuff.com</link>
	<description></description>
	<lastBuildDate>Sun, 02 Oct 2011 00:29:39 +0000</lastBuildDate>
	<generator>http://wordpress.org/?v=2.9.2</generator>
	<language>en</language>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
			<item>
		<title>Real-world data and Compare Stuff</title>
		<link>http://blog.compare-stuff.com/2007/09/27/real-world-data/</link>
		<comments>http://blog.compare-stuff.com/2007/09/27/real-world-data/#comments</comments>
		<pubDate>Thu, 27 Sep 2007 22:00:52 +0000</pubDate>
		<dc:creator>Bob</dc:creator>
				<category><![CDATA[Validation]]></category>

		<guid isPermaLink="false">http://blog.compare-stuff.com/2007/09/27/real-world-data/</guid>
		<description><![CDATA[Have you seen Swivel yet?  It&#8217;s another site for graph and data junkies like me.  You can upload data there and plot it (a handy tool for bloggers).  I find the interface a bit of a struggle when creating a new graph, but I&#8217;ve managed OK with the graph below, showing homelessness [...]]]></description>
			<content:encoded><![CDATA[<p>Have you seen <a href="http://www.swivel.com">Swivel</a> yet?  It&#8217;s another site for graph and data junkies like me.  You can upload data there and plot it (a handy tool for bloggers).  I find the interface a bit of a struggle when creating a new graph, but I&#8217;ve managed OK with the graph below, showing homelessness as a fraction of US state population.  This means we can look to see how Compare Stuff&#8217;s approach (using web search engine hit data) measures up against real-world data.<br />
<span id="more-19"></span><br />
<strong>Real world data</strong><br />
<a href="http://www.swivel.com/graphs/show/23743373"><img alt="% of State Pop. by State" src="http://www.swivel.com/graphs/image/23743373" style="border: none;" title="Click to play with this data at Swivel" /></a></p>
<p><strong>Compare Stuff web-based data</strong><br />
<a href="http://compare-stuff.com/?q1=homeless;q2=;series=160_states;y0=on;sort=1;.cgifields=y0&#038;fl=1"><img border=0 width=800 height=200 src="http://compare-stuff.com/plot.cgi?w=800&#038;h=200&#038;l=%20LA,%20WA,%20MS,%20DC,%20FL,%20GA,%20MA,%20OR,%20PA,%20VA,%20MI,%20ID,%20CT,%20AL,%20IL,%20TN,%20NM,%20VT,%20AR,%20WI,%20OH,%20NJ,%20NH,%20KS,%20NV,%20MN,%20MD,%20OK,%20CA,%20MO,%20KY,%20ME,%20AZ,%20NY,%20WV,%20UT,%20DE,%20CO,%20IN,%20TX,%20WY,%20SC,%20NC,%20ND,%20IA,%20RI,%20MT,%20AK,%20SD,%20HI,%20NE&#038;q1=0.98,0.8998,0.8811,0.8613,0.8467,0.8075,0.8047,0.7818,0.7623,0.7509,0.7363,0.7113,0.7083,0.7077,0.7038,0.6957,0.6912,0.6886,0.6632,0.6606,0.6564,0.6491,0.6475,0.6451,0.6422,0.6421,0.6388,0.6381,0.637,0.6299,0.624,0.6209,0.611,0.6064,0.6049,0.5981,0.5874,0.5815,0.575,0.5732,0.5693,0.5622,0.5567,0.5516,0.5437,0.5397,0.5255,0.5243,0.5231,0.478,0.4277&#038;q2=0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0&#038;t=bars&#038;l1=homeless&#038;l2=&#038;embedded=1"/></a></p>
<p>The top fives from each are:</p>
<ul>
<li>Real-world: District of Columbia, Nevada, Rhode Island, Hawaii, California</li>
<li>Compare-stuff: Louisiana, Washington, Mississippi, District of Columbia, Florida</li>
</ul>
<p>Not a massive overlap there (just DC).  It looks like the hurricane-hit states get more web coverage of homelessness.  Florida and Washington are ranked quite high (10 and 11) in the real-world data, but Hawaii is way off (rank 4 in real-world, rank 49 with Compare Stuff).  Sadly, there&#8217;s not necessarily a correlation between the number of homeless people in a given location and the amount of talk on the web about the subject.  Take California for example: <a href="http://compare-stuff.com/?q1=surfing;q2=;series=160_states;t.x=0;t.y=0;y0=on;sort=1;.cgifields=sort;.cgifields=y0&#038;fl=1">surfing</a>, <a href="http://compare-stuff.com/?q1=%22start-up%20company%22;q2=;series=160_states;t.x=0;t.y=0;y0=on;sort=1;.cgifields=sort;.cgifields=y0&#038;fl=1">start-ups</a> and <a href="http://compare-stuff.com/?q1=wine-making;q2=;series=160_states;t.x=0;t.y=0;y0=on;sort=1;.cgifields=sort;.cgifields=y0&#038;fl=1">wine-making</a> dominate.</p>
<p>It looks like we need an easy way to export data from Compare Stuff &#8211; to hook up with Swivel, for example, or for further offline analysis (a rank correlation coefficient would have been useful for this post, for example).</p>
]]></content:encoded>
			<wfw:commentRss>http://blog.compare-stuff.com/2007/09/27/real-world-data/feed/</wfw:commentRss>
		<slash:comments>0</slash:comments>
		</item>
	</channel>
</rss>

