Saturday, July 18, 2009

New Angus-Reid Poll: 3% Conservative Lead

Angus-Reid has released a new poll today, taken between July 16 and July 17 and involving 1,007 interviews. The national results:

Conservatives - 33%
Liberals - 30%
New Democrats - 18%
Bloc Quebecois - 11%
Greens - 6%

While it does give the Tories a lead, it actually marks a drop from 36% from the last Angus-Reid poll that was released not much more than a week ago. The Liberals have remained steady, while the NDP and Bloc get good results. The Greens are low.

Regionally, some notable results include a 29% for the NDP in the Prairies, placing them in second behind the Tories at 50%, and a small Liberal lead of 38% to 36% in Ontario. The NDP scored 17% in the province, a strong result for them. In Quebec, the Bloc is doing well with 39%, the Liberals are steady at 30%, and the Tories show a weak 13%. Atlantic Canada, which shows a Conservative lead of 37% over the NDP at 35% and the Liberals at 23%, is probably an outlier result. The Tories remain in front in BC (40%) and Alberta (63%).

This poll would result in the following seat totals:

Conservatives - 131
Liberals - 95
Bloc Quebecois - 51
New Democrats - 31

The odd Atlantic result throws things out of whack (12 Conservative seats, 11 Liberal, and 9 NDP) but as long as the Tories have a competitive result in Ontario these are the kinds of results we see.

The poll also asked about who would make the best Prime Minister. We haven't had one of those since April! So, finally, the Best PM chart has been updated. It shows Stephen Harper at 34% (up one), Michael Ignatieff at 27% (down one), and Jack Layton at 16% (up one). The actual result in the AR poll was 26% for Harper, 22% for Ignatieff, 13% for Layton, and 19% for "none of these".

The leaders were also rated on issues, and Harper came out on top for the economy (31%) and crime (34%). Layton was on top for the environment (27%) and health care (23%), and Ignatieff placed first on foreign affairs (30%).

This isn't a good poll for the Liberals, but it is a very good poll for the Conservatives, NDP, and Bloc. However, considering that the previous poll put the Conservatives ahead by six points, this is actually good news for the Liberals.

5 comments:

  1. This is really interesting information. I am glad that someone is compiling it. I will check back frequently.
    Keep up the good work!

    ReplyDelete
  2. The posting says:

    "This poll would result in the following seat totals: Conservatives - 131 Liberals - 95

    Can this really be correct?

    In the 2006 general election the Conservatives achieved a 6-point plurality over the Liberals in the popular vote (36 to 30) and yet they received only 21 more seats than the Liberals.

    But now, with this poll you are projecting a mere 3-point plurality (33 to 30) to produce 36 more seats for the Conservatives than the Liberals.

    If this is correct, it would seem to be quite a surprising result.

    ReplyDelete
  3. The national gap is not really the most important factor. It is in the regional results that the Conservatives win such a large minority.

    For example, the poor Liberal result in the Prairies (18%) results in only two seats, while the small gap in Ontario results in 43 Conservative seats and only 50 Liberal. Then the middling Quebec result gives only 19 seats and the disastrous (though likely inaccurate polling) result of 23% in Atlantic Canada gives only 11 seats to the Liberals.

    It is really in the regional results that you can figure out what a Parliament will look like.

    ReplyDelete
  4. I see your point.

    This poll result certainly illustrates how important regional numbers can be if national poll figures do not vary uniformly across the country.

    However, this highlights another matter: sample size for regional subsets.

    For example, this poll only included 1007 responses nation wide.

    This means that many of the regional subsets will have very small sample sizes.

    For example, a total of 60 or 70 people in all four Atlantic provinces combined will have been surveyed.

    A sample size of 70 can cause the margin of error to explode to more than ten per cent.

    ReplyDelete
  5. Indeed, which is why my projections for an individual poll need to be taken with a big grain of salt. It has nothing to do with the actual situation, it is just what this poll means. As the popular vote is irrelevant when it comes to forming a government, it is important to know what a polling result means in real terms.

    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.