UKPR Projection

Anthony Wells

— 2 min read
Share

Current House of Commons

Uniform Swing Projection*

UKPR Projection

mysql_connect("localhost", "ukpoll_ajwells", "haddock"); mysql_select_db ("ukpoll_pollresults");

$pollster = 'UKPR'; $avscores = mysql_query ("SELECT * FROM currentproj WHERE WHERE Pollster='$pollster'");

$conav = mysql_result ($avscores, 0, "Misc1"); $labav = mysql_result ($avscores, 0, "Misc1"); $ldav = mysql_result ($avscores, 0, "Misc1");

$conswing = $conav-33.27; $labswing = $labav-36.19; $ldswing = $ldav-22.68;

$result = mysql_query ("SELECT * FROM swingc");

$conseats = 0; $labseats = 0; $ldseats = 0; $othseats = 0;

for ( $i=0; $i < mysql_num_rows($result); $i++ ) {

/* Basic vote */

$cvote = mysql_result ( $result, $i, "Conv")+$conswing;

$lvote = mysql_result ( $result, $i, "Labv")+$labswing;

$ldvote = mysql_result ( $result, $i, "Ldv")+$ldswing;

$pvote = mysql_result ( $result, $i, "Pcv");

$svote = mysql_result ( $result, $i, "Snpv");

$ovote = mysql_result ( $result, $i, "Othv");

if (mysql_result ( $result, $i, "County")==1) {

echo "MP retiring";

$sitting = mysql_result ( $result, $i, "Sitting");

if ($sitting==1) {

$cvote=$cvote - 2; //con incumbency factor

} elseif ($sitting==2) {

$lvote=$lvote - 2; //lab incumbency factor

} elseif ($sitting==3) {

$ldvote=$ldvote - 6; //ld incumbency factor

}

} } ?>

The standard method of translating shares of the vote into seats is to use a uniform swing calculation. This means that the national change in vote share for each party is applied to each individual seat to see how that would effect the result, and then these theoretical results in each seat are totted up to produce a projection for House of Commons.

This is a crude measure and can result in some illogical and impossible projections - for example, if a poll showed Labour support dropping by 13%, as one poll did during Summer 2008, then a uniform swing calculation using those figures would project Labour getting less than zero votes in 48 seats. This is clearly nonsense. Such projections also ignore any regional variations, tactical considerations or variations due to incumbency effects of new MPs or MPs standing down. Despite all these drawbacks, it normally does a reasonably good job and, given that it is a straight extrapolation of current voting figures it is at least accepted as a fair projection that is not at the whim of individual guesswork or assumptions.

The UK Polling Report projection on the other hand IS at the whim of an individual's guesswork and assumptions - my own. At its core this is based on a uniform swing calculation, but with some additions. Firstly, I make some assumptions about the effect of personal votes and incumbency, in benefits for the incumbent in a seat where the MP was newly elected in 2005 and penalties where a sitting MP is standing down (in both cases this is especially the case for Lib Dem MPs, who enjoy higher incumbency bonuses). Secondly, I assume a differential swing in different groups of seats, both geographical and political (e.g. seats in the North or South, or Con/Lab or LD/Con). Apart from Scotland, where the swing is taken from the latest Scottish opinion polls, these differentials are based on those we found in the PoliticsHome poll of marginal seats. Thirdly, in safe seats the swing is dampened down to avoid absurd results, this results in a proportionally higher swing in marginals - though the effect is very small. Fourthly and finally, I take into account by-election results to give winning parties a boost there.

Read more