Long term readers of the blog will recall that in the run-up to the 2014 indyref, I regularly updated the Scot Goes Pop Poll of Polls, which differed from other Polls of Polls in one key respect - it only included the most recent poll from each individual polling firm. Because there were such wide disparities between the No-friendly and Yes-friendly firms, it seemed obvious to me that you'd get wildly misleading trends if you went back and forth from a sample that was 75% comprised of YouGov polls to one that was 75% comprised of ICM or Panelbase polls. I eliminated that problem by ensuring the balance between firms remained exactly the same in each update.
Since 2014, it's rarely been feasible to apply the same method, because there usually haven't been enough independence polls from a broad enough range of pollsters. But for obvious reasons, the last few weeks have seen a flurry of polls, and they've come from no fewer than six different firms - two of which (Savanta ComRes and JL Partners) had never polled on independence before. Here's what the average of the most recent poll of each of the six shows...
SCOT GOES POP POLL OF POLLS
Should Scotland be an independent country?
There's no longer the chasm between Yes-friendly and No-friendly firms that we used to see a few years ago. But for what it's worth, the most No-friendly firms at the moment are YouGov and Savanta ComRes (both of which have Yes on 53%) and the most Yes-friendly firm is Ipsos-Mori (which has Yes on 58%). Curiously, Survation are currently closer (54%) to the No-friendly end of the spectrum, even though we've always tended to regard them as a relatively Yes-friendly firm.