Quote Originally Posted by Aimless View Post
You suggested they'd go "back" to LD and BNP. The data available shows that the plurality of UKIP's voters at least claimed to have voted for Con in the previous election. The YouGov data suggest it. The ipsos-mori data and its analysis also imply the same. If you have other data that says something completely different then you're welcome to share it. If not, then you're welcome to share your revised estimates instead, along with your justification for those estimates.
My estimate is that a very large proportion of voters are former protest voters from the other protest parties. Though in 2015 the protest votes went down.

The reason for that is that if you analyse from seat to seat in 2015 the UKIP vote went up by less than what the other protest votes went down in a similar ratio. So either they're getting some of the falling protest votes ... or they're getting major party voters which are getting more than replaced by former protest voters given that major party voters were going up not down.

Finally there is a lot of evidence that voters give false-recall of how they previously voted. EG after almost every election a survey shows that more voted for the winner than actually did while the Lib Dems notoriously get far less people recalling they voted for them (their voters tend to recall voting for either of the major parties).
http://ukpollingreport.co.uk/faq-weighting
The problem with this approach is down to “false recall” – people are not actually very good at remembering how they voted at the last election. This is undisputed and supported by extremely strong evidence – multiple academic panel studies have demonstrated that if you ask a group of people how they voted, note down the results and then go back to the same people a period of time later and ask them again their answers will have changed. There are various different explanations for this phenomenon, perhaps people say how they wish they had voted, rather than how they actually did, perhaps they are embarrassed to admit voting for a now unpopular party. There is good evidence to suggest that some people align their “past vote” to match how they would vote now. My two preferred explanations are that people who did not vote claim they did (and tend to claim they voted Labour) and that Labour supporters who voted tactically for the Liberal Democrats in their own particular constituency claim they voted for the party that they actually supported, rather than the party they voted for.
Other research has shown that a significant number of people tend to exaggerate their likelihood of voting as well to report having voted when they probably could not have done so. The sample underlying that particular YouGov analysis underestimated UKIP's actual share of the popular vote by a couple of percent.
So you've demonstrated that the analysis was broken, well done.