|ComRes splits||Total %||18-24 %||25-34 %||35-44 %||45-54%||55-64 %||65 plus %|
Is this why 1 in 7 LAB supporters failed to turn-out?
One of my pet theories on opinion poll movements is that a lot of the change from survey to survey is driven by the least valuable electors of all – those who couldn’t be arsed to turn out at the preceding general election.
It was this segment, it will be recalled, which swung behind Nick Clegg’s Lib Dems in the three weeks leading to May 6th 2010 only to sit on their bottoms on the day.
The data from the overnight ComRes survey, helpfully as ever published at the same time as the poll, suggests that the same dynamic might be happening at the moment with Labour. The pollster is the only one which puts a figure on the proportion of those saying they’ll support each party who didn’t vote in the last general election.
Thus one in 23 of current LD supporters was a non-voters last May. The figure for the Tories is one in 16 while with Labour the proportion increases to one in 7.
The pollster with the best historical general election record, ICM, discounts by 50% the “value” of those sampled who were non-voters. Maybe this partly explains why the firm in recent months has consistently had Labour at lower levels.
The ComRes age break-down above shows the heavy reliance that Labour has on voters at the younger end of the age range.