Dueling polls were released in the aftermath of the BC Liberals’s throne speech. They were both conducted around the same time, but came to some different conclusions. In the week or so since they came out, I’ve been going through them and trying to figure out why their conclusions differed, and I think I may now know why.
Firstly, the polls largely agree on where the BC Greens sit. One poll (the one from Mainstreet Research) has the Greens at 17%, the same level of support they enjoyed in the recent election. The other poll (from Insights West) has the BC Greens up slightly to 19% since the election. All within the margin of error of each other, and to be expected.
The BC NDP situation varies a bunch more. Insights West has them holding their 41% total from the recent election, whereas Mainstreet has them dropping to 34%. That difference is outside the margin of error, so is very significant.
Finally, Mainstreet showing BC Liberal support increasing from its 41% on election day to 45%, while Insights West has them dropping down to 36%. That’s an extremely significant difference.
Let’s dive into the other details, and see if we can figure out why. One difference between the polls is that Mainstreet appears to have asked respondents which party they would support as the first question, whereas Insights West’s first question appears to have been what people thought about the party leaders. Andrew Weaver’s approval rating has increased to 50%, and Horgan’s has risen to 49%. Meanwhile Christy Clark’s approval rating has dropped to 32%. Only then did they ask the party support question. By focusing poll respondents first on what they thought of the party leaders, it appears to have contributed to 2% and 7% increases in the poll for the Greens and NDP respectively, and a 9% drop to the BC Liberals. With both Weaver and Horgan more popular than their parties, and Clark less popular than her’s, it’s perhaps no surprise that focusing poll respondents on their feelings on the leaders before asking what they thought of the parties would have an affect.
It’s important to note that this kind of thing is not unusual. When you hear about voting intention in opinion polls, they commonly also ask other questions and often ask those other questions first, so that doesn’t de-legitimize the Insights West poll. And when a voter goes into a ballot box, they’re voting for a party’s representative rather than a party, so the people and personalities involved are significant to the decisions of many voters.
Insights West also asked voters questions about whether people supported several policies mentioned in the BC Liberal Throne Speech, many of which were not BC Liberal policies, but were BC Green and/or NDP policies during the recent election. All of the Green and NDP policies they tested received majority support except the NDP’s plan to eliminate tolls on two Greater Vancouver bridges. Importantly to reformers, banning union and corporate donations received 80% support and holding a referendum on proportional representation found 58% support. When asked whether they thought the BC Liberals would actually actually carry through on them had the throne speech found opposition support, however, only 28% and 29% of respondents (respectively) thought they’d follow through on implementing those ideas.
So whose poll to believe? Well, for what it’s worth, Insights West absolutely nailed it in the May election. Their final pre-election poll (which included polling up to the day before the election) was 41%/41%/17%, which is close to what we got on election day, but is exactly what we got once all the absentee ballots were counted. Whether Insight West was good, lucky, or both, it was an impressive result for the pollster. So when they say that the Greens have not suffered from their alliance with the NDP (and maybe even gone up a little), that the NDP is holding steady, and that the BC Liberals are down (and that a likely reason is voter cynicism about the honesty with which they approached their failed throne speech), I’m inclined to listen to them.