Green growth in support

Home / Blogs / David Parker's blog /
By David Parker on 8 April 2010 - 12:08pm

Look at the steady growth of support for our party:

http://www.theglobeandmail.com/news/politics/ottawa-notebook/canadians-shut-door-on-afghan-combat-mission/article1527369/

It can only portent a solid base of support.

We have to hope that what happened in previous elections - change of heart at the ballot box - doesn't occur again. I think that this is a result of the ABC (Anyone But Cons) phenomenon and will happen again if the cons move into potential majority government territory. The ABC change of heart vote is relatively small but, in an despicably anti-democrat FPTP system, can change everything.

The whole scenario cries out for fair voting and we won't get it unless the opposition developes the back bone that Stefan Dion had in proposing a coalition government. The coalition was a revolutionary idea whose time is long past and reflects the true nature of the democratic idea - rule by the largest number of voter and a close reflection of true voter wishes. A 40% support base should, in an ideal world, put a party in opposition.

Coalitions governments are inherently more conservative in a strange way. A party must compromise to rule and any deviation in policy that is opposed by the majority will fail without full coalition support. Were this to have happened in the past we wouldn't have the Free Trade Agreement, nor the GST, we certainly wouldn't have even considered Iraq and might not have gone into Afganistan. Whether you like these policies or not is irrelevant because they must meet a concensus vote in a true democracy or fail.

Let's not allow the continued ignorance of the electoral system that is most common in the developed world to prevent us from promoting it even more.

 

Comments

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Only one poll counts

Submitted by John Northey on 8 April 2010 - 3:56pm.

I thought it would be worthwhile to see what happened pre-2008 and see how we did in polls then.  Wikipedia has it here.

We ranged from 4% (Environics 6/23/2006) to 14% (Harris-Decima 9/18/2007).  12 out of 300 polls did not have us listed (Decima Research mainly with the final not listed being on 4/21/2008).  The average was 9.1% vs the 6.8% we got.  Only the Liberals dropped vs their overall average as well, by 2.4%.  Only 28 polls had us under 6.8%, while 259 had us above.

Most of the 'low' polls were in 2006 but 5 were in October just before the election.  Those were from Angus Reid Strategies (2), Nanos Research (2), and EKOS.

Worth noting.  I'll dig deeper when I have time.

John Northey
Wellington-Halton Hills

Poll Results by Polling Firm

Submitted by David Barclay on 10 April 2010 - 10:01am.

Since each polling company uses a different methodology, mainly what they theorize as a ‘representative sample’. It may be misleading to aggregate all the results on one plot.  Plotting major individual firms’ trendlines:

Ekos:                        positive   (but huge gap of over 15 months)

Angus:                      negative

Harris:                      negative

Ipsos:                        positive

Nanos:                      negative

Strategic Cnl:            positive

       Unfortunately there is no distinct trend emerging here either. (This is derived from the same data set as below.)

Respectfully, D. Scott Barclay

Poll Results vs. a Wish and a Hope

Submitted by David Barclay on 10 April 2010 - 9:46am.

There seems to be a fascination with wildly subjective interpretation of poll results throughout the blogs, so let’s try to apply standard statistical analysis. (Forgetting that new results can turn on a dime by a major political event.)

Source of data:  http://www.electionalmanac.com/canada/polls.php

From:  2006.02.09  to  2010.02.09.  Poll results where GPC was not included were eliminated from data set.  Total of 394 poll results. The most recent polls were not included, as I could not find all the representative polling firms.

Sources of plotting error:  MS Excel cannot recognize nor convert a numerical date of the form: 2009.06.25 into a standard date/time value. (If someone can create a formula to do this pls let me know. Changing it into a number like 20,090,625 does not represent the time value.)

Therefore for plotting purposes I converted it to number of months since Jan.1/06. The effect of this then is to lump all the results for each month together. So any trend is based on month-to-month without recognizing the day of month the poll was taken on.

Results:   Range (extremes) of y-axis was 3% in 2009 to 14% in 2007)   Each year’s results were spread over quite a wide range:  2006: 4 - 11%   2007: 6 - 14%   2008: 6 – 13%   2009: 3 - 11.6 %   2010: 5.6 – 13.4% (2 months only)

Regression Analysis assuming;

Linear (straight line):  8.2 to 9.6%  for a 1.4% increase since Feb. 2006, or  0.35% per 12 months.

Exponential:  7.95 to 9.25%  for a 1.3% increase since Feb. 2006

(Considering only the data from Jan. 2008 until Feb. 2010, the linear trend is actually downward from 9.65 to 8.8%)

I’m pretty rusty at this, so if there is someone who can improve and extract more representative results from the data, feel free. I can send anyone the plot and XL spreadsheet if they like.  The long-term trend is positive. We can speculate all we like, but without a disruptive event, or a new ‘game-changing strategy’ by the GPC, there is no indication that there is anything but a mild upward trend.

 

 

Respectfully, D. Scott Barclay