Thursday, September 15, 2011

Federal status quo in this week's Nanos

On Monday, Nanos Research released its latest federal numbers for The Globe and Mail, finding that the Conservatives and New Democrats are on the up-swing at the expense of the Liberals. But in the end, all we really get is the status quo.
Compared to Nanos's last poll taken at the end of July and the beginning of August, the Conservatives have gained 3.2 points and lead with 39.5% support. They had 39.6% on election night.

The New Democrats are up 6.3 points to 33.1%, while the Liberals are down 6.3 points to 20.7%.

Though this indicates real movement, the Nanos poll that had the Liberals and NDP tied was different from every other poll. So, in all likelihood, this Nanos poll is more of a reset.

The Greens have dropped 1.1 points to 3.4% while the Bloc Québécois is down 1.8 points to 2.6%.

In Ontario, the parties are generally where they stood on May 2nd with a slight increase for the Conservatives. Compared to their earlier poll, the Tories are up nine points to 46.9%, well ahead of the Liberals (26.7%, down 5.5 points) and the NDP (24.4%, down 0.4 points). This alone more or less guarantees a Conservative majority.

Nanos has the NDP dominant again in Quebec, up 14.7 points to 48.9%. The Conservatives drop 5.4 points to 18.8%, while the Liberals are down 6.1 points to 16.1%. The Bloc is down 7.7 points to 9.6%, a miniscule amount of support. Note, however, that Nanos has averaged 12.9% for the Bloc since the election, compared to 24.3% in the just-as-numerous Abacus polls, while Léger, Crop, Harris-Decima, and Angus-Reid have all had the Bloc at 18% or more. Nanos seems to be the dissenting opinion on Bloc support in Quebec.

The New Democrats are leading in British Columbia with 37.6% (up 7.8 points), while the Conservatives are close behind with 36.7% (up 3.9 points). The Liberals trail with 21.4% and the Greens have sunk to only 4.3% support.

The Tories comfortably lead in Alberta, Saskatchewan, and Manitoba, as is to be expected.

This poll would not result in a very different House of Commons. The Conservatives win 164 seats, the New Democrats 110, and the Liberals 34.

The Bloc Québécois is completely shut-out, while Elizabeth May is not re-elected.

The Conservatives win one seat in the north, 17 in British Columbia, 27 in Alberta, 17 in the Prairies, 75 in Ontario, five in Quebec, and 22 in Atlantic Canada. Their losses in the West are made up in the East.

The New Democrats win one seat in the north, 14 seats in British Columbia, one in Alberta, six in the Prairies, 20 in Ontario, 62 in Quebec, and six in Atlantic Canada.

The Liberals win five seats in British Columbia, five in the Prairies, 11 in Ontario, eight in Quebec, and four in Atlantic Canada.

This poll was taken the week after the Layton funeral, so emotions might have still been running high. But we've seen elsewhere that the New Democrats are still doing very well in Quebec and appear to be in a very good position in British Columbia. Maintaining this level of support out West could be an important component of a future NDP government.

The only problem is, as it is always is for any party, Ontario. The New Democrats are far from a breakthrough in the province, but it is absolutely essential that they win another 20 seats or so in the province if they want to have a hope of winning more seats than the Conservatives nationally, as it is difficult to see where else their new seats could be won.

9 comments:

  1. I wish we knew what 24% support for the NDP in everything-between-BC-and-Ontario actually translated to. Is it a relatively useless bubble in Alberta? Or is it a real, and significant, increase of support in more fertile NDP territory to the east?

    ReplyDelete
  2. A question for you. Why do you keep comparing a pollster's current poll to actual election results on May 2, 2011, to try and show trends and which party is gaining and which party is losing support? Comparing poll results to actual election results is like comparing apples to oranges. The two are completely different in methodology and sampling.

    The more accurate comparison would be to compare Nanos' current poll to his last poll just prior to the election - if you want to show trends and which party is rising and which is declining, post election.

    Assuming Nanos is using the same polling methodology and sampling, it is appropriate to compare poll to poll, not poll to election results which are entirely different.

    Nanos' last poll before the election substantially underestimated the Conservatives (as all pollsters did). I'm not saying his polling is biased, just that his methodolgy underestimated actual Conservative strength. Therefore, you will get a misleading trend if you compare his current poll to election results - assuming he is using the same polling methodology and sampling. Comparing poll to poll shows the Conservatives gaining strength since the election. Comparing poll to election results shows the Conservatives remaining static.

    Again, I see no rational in comparing any poll from any pollster to election results to show a trend. Instead, the proper comparison is to compare current polls to the pollster's last poll just prior to the election. If you disagree please explain. Thanks.

    ReplyDelete
  3. A poll is a snapshot in time and has a margin of error.

    Nanos didn't poll on election night, so what you are arguing is that I should say that Conservative support has increased since the weekend before the election (when Nanos last polled). Not sure why that would be relevant.

    ReplyDelete
  4. Anonymous 14:54, you are assuming that whatever fault resulted in underestimated Conservative support is persisting, and you are doing so without any evidence. Your logic is faulty.

    If you actually read for comprehension, you would notice that when describing trends, Eric did talk about change from the last Nanos poll rather than from the last election. The only reference to the last election was saying that the numbers that Nanos produced were about the same as the numbers from May 2. No conclusions were drawn based on that.

    Pull your claws in and read more carefully.

    ReplyDelete
  5. Anonymous 14:54,

    "Nanos' last poll before the election substantially underestimated the Conservatives (as all pollsters did)" - Perhaps you should check your facts. Nanos was (the only pollster) within the MOE for all parties.

    http://en.wikipedia.org/wiki/Opinion_polling_in_the_Canadian_federal_election,_2011

    http://www.nanosresearch.com/election2011/Closeout-Memo-2011E.pdf

    ReplyDelete
  6. TS, well said. "Pull your claws in" is wonderful advice, whether online or in the real world. So much reflexive hostility in the commenting.

    Éric, I would be interested to hear your reaction to what the Globe and Mail is describing as a row between Ipsos and Abacus. http://www.theglobeandmail.com/news/politics/canadian-pollsters-row-shows-there-is-no-safety-in-numbers/article2168246/

    Thanks!

    ReplyDelete
  7. I think both sides make good points, but I can't see how it helps the industry. For many people not interested in the details, they'll just look at the headline and tar everyone with the same brush.

    ReplyDelete
  8. And don't anybody make the mistake of equating Federal support in Ontario for Provincial support for any party. It don't work that way !!

    ReplyDelete
  9. Most polls tend to underestimate Tory support. I think the Tories are gong to be in power for some time as the Liberals are damaged goods and the NDP is highly unlikely to do any better than they did on 2 May. Jack Layon did the whole of Canada a massive favour exterminating the Bloc and slapping down the Liberals. Jack Layton would have made a great Tory.

    ReplyDelete

COMMENT MODERATION POLICY - Please be respectful when commenting. If choosing to remain anonymous, please sign your comment with some sort of pseudonym to avoid confusion. Please do not use any derogatory terms for fellow commenters, parties, or politicians. Inflammatory and overly partisan comments will not be posted. PLEASE KEEP DISCUSSION ON TOPIC.