The election day to be announced with Labour in trouble
Anthony Wells gives us the low-down on FOUR polls in today's papers all of which show Labour's lead collapsing, or in the case of Mori's FT poll, a 5% TORY LEAD.There are some important caveats. This is only one poll (though the other three aren't, obviously). The poll could be wrong. The margin of error (which Mori don't give oddly) is about +_4% with this sample size (so the Tories, rogue polls notwithstanding, are in the lead).
Most importantly though the poll is markedly sensitive to changes in turnout -- it forecasts one of 55%, at 60% the gap is 3%, at 70% the parties are level-pegging. At 90% Labou are 5% points ahead. As Bob Worcester notes 70% was the minimum turnout at every post-war election until 2001 (though of course political participation has declined a lot generally since the 1970s). All-in-all the 45% who are not certain to vote support Labour by about 10% points ahead of the Tories (this is what Worcester says, I think it's more like 20%).
Nevertheless I can't remember the last time a PM with a large majority (in this case of over 150) and a full-year to go called an election with his party's poll standing so weak. So why is he risking it?
posted by Matthew @ 4/05/2005 06:52:00 AM
Comments: 0
