Home FAQ Search Memberlist User Groups Register Login   

BloggingTories.ca Forum IndexBloggingTories.ca Forum Index
    Index     FAQ     Search     Register     Login         JOIN THE DISCUSSION - CLICK HERE      

*NEW* Login or register using your Facebook account.

Not a member? Join the fastest growing conservative community!
Membership is free and takes 15 seconds


CLICK HERE or use Facebook to login or register ----> Connect



Goto page Previous  1, 2, 3, 4  

Post new topic   Reply to topic Page 4 of 4
View previous topic :: View next topic  
Author Message
cosmostein





Joined: 04 Oct 2006
Posts: 7628
Reputation: 305.3Reputation: 305.3
votes: 21
Location: The World

PostPosted: Mon Nov 13, 2017 3:47 pm    Post subject: Reply with quote

The Poll for anyone interested;
http://poll.forumresearch.com/.....vember.pdf

While it has the Tories ahead it still has the LPC ahead on seat count. (148 Vs 164)

Its promising that the Tories are in the 30s on Atlantic Canada;
Its also nice to see a lead that goes Ontario all the way West (All be it the lead in Ontario is 1).

The concern I have is the NDP numbers;
I don't believe they are 30 points behind the CPC in BC and I am not sure if I buy the NDP @ 15 in Ontario.
RCO





Joined: 02 Mar 2009
Posts: 7245
Reputation: 252.6
votes: 3
Location: Ontario

PostPosted: Mon Nov 13, 2017 4:11 pm    Post subject: Reply with quote

cosmostein wrote:
The Poll for anyone interested;
http://poll.forumresearch.com/.....vember.pdf

While it has the Tories ahead it still has the LPC ahead on seat count. (148 Vs 164)

Its promising that the Tories are in the 30s on Atlantic Canada;
Its also nice to see a lead that goes Ontario all the way West (All be it the lead in Ontario is 1).

The concern I have is the NDP numbers;
I don't believe they are 30 points behind the CPC in BC and I am not sure if I buy the NDP @ 15 in Ontario.



the poll does have some odd numbers , has the cpc at 74 % in man/sask , no doubt there leading but that seemed high .

but in general there numbers are in line with other polls , liberals leading in atlantic Canada and Quebec . cpc leading in alberta and western Canada , with Ontario close to tied

they also show the ndp fighting with the greens in some regions for 3rd place , have the greens slightly ahead of the ndp in atlantic Canada and in BC have them both at 12 % . an election at this time would be a nightmare for the ndp as there not doing good anywhere and have no regional strength even in places they hold seats

but 15 % in Ontario seems reasonable as the federally ndp have been polling less than provincial ndp which I believe is in mid 20% , so that would seem reasonable to me ,
cosmostein





Joined: 04 Oct 2006
Posts: 7628
Reputation: 305.3Reputation: 305.3
votes: 21
Location: The World

PostPosted: Thu Jan 18, 2018 1:15 pm    Post subject: Reply with quote

Anyone who following the Nanos Weekly tracker has seen their rolling methodology, basically my understanding is the numbers are based on a four week rolling average of 1000 respondents (or 250 per week over that four week period)

Generally speaking it takes a while to reflect the current feeling as you have information from four weeks prior still being factored in.

This weeks was a bit of a head scratcher for me;
http://www.nanosresearch.com/t.....%20Eng.pdf

The week prior the LPC had a 10.2% lead over the CPC;
This week its reduced to 3.2%

If I am understanding the methodology correctly all that changed was the 250 sample from the week prior to December 15, 2017 dropped off but an addition of 250 new samples to the pool dropping the lead 7% seems massive?
Bugs





Joined: 16 Dec 2009
Posts: 4676
Reputation: 253.8
votes: 8

PostPosted: Thu Jan 18, 2018 1:54 pm    Post subject: Reply with quote

The merit of these rolling polls is that they are more current than the normal polls. This works fairly well if the 250 participants are selected daily, but if they are selected a week apart it loses the point. It amounts to a series of weekly, separate polls with a sample size of 250 -- it will have a huge margin of error. so big as to be practically meaningless.
cosmostein





Joined: 04 Oct 2006
Posts: 7628
Reputation: 305.3Reputation: 305.3
votes: 21
Location: The World

PostPosted: Thu Jan 18, 2018 3:48 pm    Post subject: Reply with quote

Bugs wrote:
The merit of these rolling polls is that they are more current than the normal polls. This works fairly well if the 250 participants are selected daily, but if they are selected a week apart it loses the point. It amounts to a series of weekly, separate polls with a sample size of 250 -- it will have a huge margin of error. so big as to be practically meaningless.


Agreed,
But in a pool of 1000 samples if you drop 250 of the oldest and add 250 of the newest and you have a 7 point difference in the gap between the top two parties,

Either the oldest sample was incredibly in favor of the Liberals and the latest have been closer, or the new sample is incredibly in favor of the CPC with the older ones being more spread.
Bugs





Joined: 16 Dec 2009
Posts: 4676
Reputation: 253.8
votes: 8

PostPosted: Sat Jan 20, 2018 12:00 pm    Post subject: Reply with quote

True but if you have a total sample size of 1000, and you roll over 250 every week, the opinions you are dropping are amost five weeks old (from the time the data was collected) ... a poll result published today would have a quarter of its data collected last year!

So the current poll results include data from the post-Christmas warm glow with the colder feelings about how the trade negotiations are going and such stuff.
Post new topic   Reply to topic Page 4 of 4

Goto page Previous  1, 2, 3, 4  


 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum


Conservatives would beat Libs: Poll

phpBBCopyright 2001, 2005 phpBB