It’s generally forgotten now, but five or so years ago there was a major controversy in British polling about whether to prompt the option of UKIP or not.
Prompting is where an option is presented up-front by a pollster, rather than having to be specified by the respondent. So while nowadays polls will prompt UKIP by asking “Who would you vote for in a General Election tomorrow – Conservative, Labour, Liberal Democrat, UKIP, or other?”, back then the party was un-prompted, meaning the options would be “Conservative, Labour, Liberal Democrat, or other”, and respondents would have to say “Other” then choose UKIP either by specifying it themselves or picking the party from a secondary list of minor parties.
Not prompting a party lowers its poll-rating, for the obvious reason that it makes it harder and therefore less likely for someone to say they plan to vote for them. That was a reasonable decision back when UKIP were still fairly obscure, and weren’t fielding candidates in every seat – prompting them too early would have artificially inflated their poll performance above their actual vote. But as the Farage surge took off in a series of by-elections and then the 2014 EU election, the situation reversed – UKIP argued that the failure to prompt was understating their support given that they had become a truly national party.
Eventually, various pollsters came to accept that criticism, and added the party to their standard list of prompted options (some, it should be noted, don’t prompt at all, while others don’t make a distinction). That in itself helped to increase the perception of a surge, producing a feedback loop of sorts, and arguably led to some overstatement of their vote, but it was certainly merited and more fair than the previous approach, which had become outdated.
We’re now reaching a point, however, where it is reasonable to ask if UKIP should revert to being unprompted in the polls.
They have lost every single council seat they have sought to defend for at least the last year, and received fewer votes at the 2017 General Election than at any national ballot since 2001. Since then, their membership has fallen severely, they have gained then lost yet another leader, and the Party is now reported to be on the brink of bankruptcy. On Monday, UKIP was ordered to pay £175,000 in court costs in fairly short order, which poses a real threat of finishing them off as an organisation.
Don’t get me wrong; there are still UKIP voters and activists around the country. There probably will be for a very long time, even if the central setup has gone (the SDP only lost its last council seat in 2014, for example). But it’s increasingly hard to justify continuing to include the party as a prompted national polling option. If it had been in this dire state back in 2014, there’s no way anyone would have agreed to prompt it. Given that it currently looks set to either collapse entirely or fail to put up candidates in most seats, its continued prompting should surely be in doubt.
This raises an interesting question. While UKIP has nosedived in the polls, it still retains a few percentage points – somewhere between one and four per cent in recent weeks, according to UK Polling Report. If the pollsters do stop prompting it, where might this vote go? Labour and the Conservatives are still very close, each at or just above 40 per cent, and even small amounts of additional support could make quite a big perception and psychological difference to one or both sides. I’d be surprised if the major pollsters weren’t already testing exactly this possibility to study what the effects might be on their modelling.