Labour’s collapse in the polls is NOT because of a fall-off in Lib Dem switching
The above monthly averages are based on analysis of data from every single Populus online poll since the firm started its twice weekly series in July 2013 and shows the proportion, before the don’t knows are netted off, of 2010 LDs now saying they will vote Labour.
Because of the way Populus presents its data this is a much easier number gathering exercise than it would be with YouGov which only shows the netted off numbers in its datasets. As a result of these Populus figures being monthly averages they are based on serious overall sample sizes – about 15k – so we can view them with more confidence. Just taking something from one poll can be misleading which is why I’ve approached this task in this way.
As can be seen the trend has been relatively consistent although there was a down-turn in October but not to a low point for the series.
Clearly the LAB polling position has been hit by Scotland but that can only account for a maximum of 1% to 2%.
What we’ve seen is a general drift downwards which seems to have been triggered by more switching to UKIP, and a higher level of former voters saying don’t know.
I used to think that almost whatever happened the 2010 LD switchers, if they remained, would give Ed a lead on seats if nothing else next May. Now I’m not so sure.