<?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: Nerdy Sunday: Predicting campaign effects</title>
	<atom:link href="http://blogs.crikey.com.au/pollytics/2009/02/15/nerdy-sunday-predicting-campaign-effects/feed/" rel="self" type="application/rss+xml" />
	<link>http://blogs.crikey.com.au/pollytics/2009/02/15/nerdy-sunday-predicting-campaign-effects/</link>
	<description>Politics, elections and piffle plinking</description>
	<lastBuildDate>Mon, 23 Nov 2009 10:16:08 +1100</lastBuildDate>
	<generator>http://wordpress.org/?v=2.8.6</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Kevin Bonham</title>
		<link>http://blogs.crikey.com.au/pollytics/2009/02/15/nerdy-sunday-predicting-campaign-effects/comment-page-1/#comment-12155</link>
		<dc:creator>Kevin Bonham</dc:creator>
		<pubDate>Tue, 17 Feb 2009 14:38:19 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.crikey.com.au/pollytics/?p=3893#comment-12155</guid>
		<description>I notice that in five of the six cases there is a movement from poll at campaign start to election result that is in favour of the incumbent government at the start of the campaign.  The exception is 2001 in which the incumbent government got a huge spike around the start of the campaign from Tampa/S11, which then washed out of the system by election day.  

Also in four of the five cases other than 2001, the incumbent government trailed in the polls at the start of the campaign, while in the fifth it was virtually line-ball.  So are we actually seeing anything more here than the usual pattern for the government&#039;s position to appear worse than it is in early polling, and a drift towards the government as undecideds stick with what they know therefore often creating a trend towards the previous election result?  Would be nice to have several dozen data points instead of just six - throw in the more reliably polled state elections maybe?</description>
		<content:encoded><![CDATA[<p>I notice that in five of the six cases there is a movement from poll at campaign start to election result that is in favour of the incumbent government at the start of the campaign.  The exception is 2001 in which the incumbent government got a huge spike around the start of the campaign from Tampa/S11, which then washed out of the system by election day.  </p>
<p>Also in four of the five cases other than 2001, the incumbent government trailed in the polls at the start of the campaign, while in the fifth it was virtually line-ball.  So are we actually seeing anything more here than the usual pattern for the government&#8217;s position to appear worse than it is in early polling, and a drift towards the government as undecideds stick with what they know therefore often creating a trend towards the previous election result?  Would be nice to have several dozen data points instead of just six &#8211; throw in the more reliably polled state elections maybe?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Possum</title>
		<link>http://blogs.crikey.com.au/pollytics/2009/02/15/nerdy-sunday-predicting-campaign-effects/comment-page-1/#comment-12128</link>
		<dc:creator>Possum</dc:creator>
		<pubDate>Mon, 16 Feb 2009 11:21:30 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.crikey.com.au/pollytics/?p=3893#comment-12128</guid>
		<description>fmark,

The 1998 election was a bit of a rogue event for just about every political pattern, or trend, or swing, or demographic analysis you could run because of the way the large One Nation vote affected the two party preferred vote.

One Nation was a disorganised rabble and their how to vote cards (when they existed) had no real meaning or pattern to them from one electorate to the next.

As a result, seats swung wildly in ways that often didn&#039;t make much sense. It wasnt until after the 2001 election that things got back to normal.

I often just leave the 1998 election result completely out of longer term electoral analysis. Large third parties can wreak havoc on the tpp vote - 1998 is a really good case in point.</description>
		<content:encoded><![CDATA[<p>fmark,</p>
<p>The 1998 election was a bit of a rogue event for just about every political pattern, or trend, or swing, or demographic analysis you could run because of the way the large One Nation vote affected the two party preferred vote.</p>
<p>One Nation was a disorganised rabble and their how to vote cards (when they existed) had no real meaning or pattern to them from one electorate to the next.</p>
<p>As a result, seats swung wildly in ways that often didn&#8217;t make much sense. It wasnt until after the 2001 election that things got back to normal.</p>
<p>I often just leave the 1998 election result completely out of longer term electoral analysis. Large third parties can wreak havoc on the tpp vote &#8211; 1998 is a really good case in point.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: fmark</title>
		<link>http://blogs.crikey.com.au/pollytics/2009/02/15/nerdy-sunday-predicting-campaign-effects/comment-page-1/#comment-12098</link>
		<dc:creator>fmark</dc:creator>
		<pubDate>Sun, 15 Feb 2009 23:15:54 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.crikey.com.au/pollytics/?p=3893#comment-12098</guid>
		<description>Thanks Poss, very informative.  An argument in favour of Reversion to Trend (albeit with one one data point!) winning out in 2010 is the importance of individual politicians rather than just parties.

It&#039;s interesting to note that the Reversion to Election effect is smallest in 1998, the election after PJK lost, giving only 0.5% to the incumbent.  Are we likely to see a similarly small Reversion to Election effect, given the possibility that a number of voters were voting against JWH in 2007, not the Liberals?</description>
		<content:encoded><![CDATA[<p>Thanks Poss, very informative.  An argument in favour of Reversion to Trend (albeit with one one data point!) winning out in 2010 is the importance of individual politicians rather than just parties.</p>
<p>It&#8217;s interesting to note that the Reversion to Election effect is smallest in 1998, the election after PJK lost, giving only 0.5% to the incumbent.  Are we likely to see a similarly small Reversion to Election effect, given the possibility that a number of voters were voting against JWH in 2007, not the Liberals?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Possum</title>
		<link>http://blogs.crikey.com.au/pollytics/2009/02/15/nerdy-sunday-predicting-campaign-effects/comment-page-1/#comment-12096</link>
		<dc:creator>Possum</dc:creator>
		<pubDate>Sun, 15 Feb 2009 09:38:30 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.crikey.com.au/pollytics/?p=3893#comment-12096</guid>
		<description>Thanks DR.

To be brutally honest, being a Lib at the moment is staring into a very, very hard slog.

Just about every metric or relationship or voting pattern one looks at, the Libs are on the wrong side of it at the moment. They&#039;ll be earning their keep if nothing else.</description>
		<content:encoded><![CDATA[<p>Thanks DR.</p>
<p>To be brutally honest, being a Lib at the moment is staring into a very, very hard slog.</p>
<p>Just about every metric or relationship or voting pattern one looks at, the Libs are on the wrong side of it at the moment. They&#8217;ll be earning their keep if nothing else.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: David Richards</title>
		<link>http://blogs.crikey.com.au/pollytics/2009/02/15/nerdy-sunday-predicting-campaign-effects/comment-page-1/#comment-12095</link>
		<dc:creator>David Richards</dc:creator>
		<pubDate>Sun, 15 Feb 2009 06:11:42 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.crikey.com.au/pollytics/?p=3893#comment-12095</guid>
		<description>typically great work, poss!

Neither effect would be very comforting to the Libs... they would need a massive and unprecedented change to have a sniff

Playing musical chairs with the shadow treasury position, and quite possibly another leadership change aren&#039;t likely to help either

If I was a Lib - I&#039;d be finding excuses to retire right about now.</description>
		<content:encoded><![CDATA[<p>typically great work, poss!</p>
<p>Neither effect would be very comforting to the Libs&#8230; they would need a massive and unprecedented change to have a sniff</p>
<p>Playing musical chairs with the shadow treasury position, and quite possibly another leadership change aren&#8217;t likely to help either</p>
<p>If I was a Lib &#8211; I&#8217;d be finding excuses to retire right about now.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Possum</title>
		<link>http://blogs.crikey.com.au/pollytics/2009/02/15/nerdy-sunday-predicting-campaign-effects/comment-page-1/#comment-12094</link>
		<dc:creator>Possum</dc:creator>
		<pubDate>Sun, 15 Feb 2009 04:39:24 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.crikey.com.au/pollytics/?p=3893#comment-12094</guid>
		<description>Aw shucks Don, you&#039;ll make a Possum blush :-D</description>
		<content:encoded><![CDATA[<p>Aw shucks Don, you&#8217;ll make a Possum blush <img src='http://blogs.crikey.com.au/pollytics/wp-content/mu-plugins/tango-smilies/tango/face-smile-big.png' alt=':-D' class='wp-smiley' /> </p>
]]></content:encoded>
	</item>
	<item>
		<title>By: don</title>
		<link>http://blogs.crikey.com.au/pollytics/2009/02/15/nerdy-sunday-predicting-campaign-effects/comment-page-1/#comment-12093</link>
		<dc:creator>don</dc:creator>
		<pubDate>Sun, 15 Feb 2009 04:21:07 +0000</pubDate>
		<guid isPermaLink="false">http://blogs.crikey.com.au/pollytics/?p=3893#comment-12093</guid>
		<description>Poss, you are amazing. 

I am not politically or intellectually astute enough to follow that, maybe I need another red or something, but wow, you really know your stuff. 

Here&#039;s mud in your eye, and power to your elbow. 

Don</description>
		<content:encoded><![CDATA[<p>Poss, you are amazing. </p>
<p>I am not politically or intellectually astute enough to follow that, maybe I need another red or something, but wow, you really know your stuff. </p>
<p>Here&#8217;s mud in your eye, and power to your elbow. </p>
<p>Don</p>
]]></content:encoded>
	</item>
</channel>
</rss>
