A significant development in Canadian politics has taken place over the past 24 hours — former Liberal MP, now an independent, Blair Wilson, has joined the Green Party, thus giving the Greens a sitting MP in parliament and ensuring Elizabeth May’s participation in the Leader’s Debate. Stay tuned…
A poll of Guelph’s by-election was recently conducted by an up-start Winnipeg-based polling firm KLR VU run by Allan Bruinooge, brother of Conservative MP Rod Bruinooge.
KLR VU Poll (through 13 Aug)
Frank Valeriote, Liberal — 34.6 to 38.6%
Gloria Kovach, Conservative — 24.4 to 28.4%
Mike Nagy, Green — 17.2 to 21.2%
Tom King, NDP — 15.8 to 19.8%
The above ranges reflect the margin of error of the poll. KLR VU surveyed 3,396 people on two days, July 27 and August 13, which represents a margin of error of +/- 2%, 19 times out of 20 (note: typically a survey of 3,396 people yields a margin of error of +/- 1.7%, so we assume KLR VU is rounding this to 2%).
The poll has sparked controversy due to its methodology, its perceived Conservative connection and also because it’s only other public poll — on whether Dr. He.nry Morgentaler should receive the Order of Canada or not — showed very different results from the mainstream pollsters (Ipsos and Angus Reid). David Akin has the best run-down of the controversy.
On the question of methodology, we are less concerned about the use of touchtone polling in general — where questions are pre-recorded, voters are called randomly, and asked to press ’1′ or ’2′ etc in response to different questions. No doubt this type of survey has some selection bias in that some people will simply hang up because there is no live person on the other end. More importantly, since there is no screening, it does not ensure that the respondent is an actual eligible voter (although realistically there is no way to ensure this with traditional polling either — essentially we have to take the person’s word for it). The selection bias is probably not that significantly different than traditional polling (a live person asking the questions), where people can also refuse to participate. At issue is whether supporters of one party/candidate are more or less likely to not participate than others. It seems unlikely — that is, Conservative voters are probably just as likely to hang up as Liberal voters as NDP voters and so on. Likewise, are people who are not eligible voters more likely to participate using touchtone polling than traditional polling? Probably not. So, we’re not prepared to discount the use of touchtone polling in general.
Somewhat more concerning in this poll is the large time gap between the initial survey and the final survey — 17 days apart. Given that polls are snapshots in time, a 2+ week gap certainly raises eyebrows at to its reliability. Typically, polls are most reliable when they are conducted over a short, contiguous period of 2-3 days. However, it’s probably worth noting that another mainstream pollster, Environics, also often conducts polls over a large period. For example, it’s latest survey was conducted June 13 to July 5, a gap of 22 days (although it is probable the surveys were more evenly distributed than the KLR VU poll). Given that Environics is an established pollster, the time gap in the KLR VU poll may not be significantly out of line with the mainstream, although it is not good polling practice.
Many NDP supporters, in particular, have questioned the poll’s validity, no doubt because the results show its candidate running behind the Green Party candidate. The NDP views Tom King as a star candidate and believes it can win the riding. However, it should come as no surprise to see the Greens doing well in Guelph — Guelph Greens won more than twice the province-wide Green share in the last federal and provincial elections (indeed out-polled the NDP in the latter), and the Greens are putting serious money and bodies into the riding. Compare the KLR VU results with our analysis from 14 August:
DemocraticSPACE Estimate (through 14 Aug)
Frank Valeriote, Liberal — 28-32%
Gloria Kovach, Conservative — 26-30%
Tom King, NDP — 20-24%
Mike Nagy, Green — 18-22%
The KLR VU results are slightly lower than our estimate for the NDP (the high end of the KLR VU range of 19.8% is just below our low end of 20%) and higher than our estimate for the Liberals (the low end of the KLR VU of 34.6% is just above our high end of 32%), but are within our estimated ranges for the Conservatives and Greens. So the results are not wildly different from our estimated ranges.
The jury is of course still out on KLR VU, given that it just doesn’t have a track record for us to judge its reliability. However, the results as reported are not completely out of line with what we might expect to see, plus or minus a couple of points for each party.
It’s important to keep in mind that this is, however, just one poll (although with a large sample size). Only with a few polls from different firms can we be more confident in the polling numbers. So stay tuned…
Last updated: Aug 18, 2008
If the Presidential election was held today, Barack Obama would win — 275 electoral college votes to John McCain’s 263. Here is the latest state-by-state breakdown (sorted from most to least electoral college votes). Note: the number in parenthesis is the percentage lead the candidate is ahead and EV means electoral college votes. Top battleground states (where margin-of-victory is 5% or less) are highlighted in red.
Obama beats in McCain in:
California (+15.3) — 55 EV
New York (+16.0) — 31 EV
Illinois (+21.7) — 21 EV
Pennsylvania (+6.8) — 21 EV
Michigan (+3.2) — 17 EV
New Jersey (+10.7) — 15 EV
Massachusetts (+12.6) — 12 EV
Washington (+10.5) — 11 EV
Indiana (+0.5) — 11 EV
Minnesota (+2.6) — 10 EV
Wisconsin (+7.2) — 10 EV
Connecticut (+18.6) — 7 EV
Oregon (+7.0) — 7 EV
Iowa (+5.7) — 7 EV
New Mexico (+5.2) — 5 EV
New Hampshire (+1.4) — 4 EV
Maine (+15.0) — 4 EV
Rhode Island (+22.4) — 4 EV
–
TOTAL — Barack Obama 275 EV
McCain beats in Obama in:
Texas (+10.5) — 34 EV
Florida (+1.8) — 27 EV
Ohio (+1.5) — 20 EV
North Carolina (+4.3) — 15 EV
Georgia (+6.3) — 15 EV
Virginia (+0.6) — 13 EV
Missouri (+2.3) — 11 EV
Tennessee (+15.3) — 11 EV
Arizona (+13.7) — 10 EV
Colorado (+0.5) — 9 EV
Louisiana (+16.7) — 9 EV
Alabama (+21.6) — 9 EV
Kentucky (+16.0) — 8 EV
South Carolina (+9.4) — 8 EV
Oklahoma (+29.0) — 7 EV
Mississippi (+11.3) — 6 EV
Kansas (+11.7) — 6 EV
Arkansas (+16.3) — 6 EV
Nevada (+2.4) — 5 EV
Utah (+23.3) — 5 EV
Nebraska (+18.7) — 5 EV
West Virginia (+13.0) — 5 EV
Idaho (+14.5) — 4 EV
Montana (+5.3) — 3 EV
Wyoming (+16.0) — 3 EV
North Dakota (+1.0) — 3 EV
South Dakota (+8.8) — 3 EV
Alaska (+7.0) — 3 EV
—
TOTAL – John McCain 263 EV
Note
Unpolled safe Democrat states (23 EV)
Maryland (10), Hawaii (4), Vermont (3), Delaware (3), and DC (3).
Earlier, we took a look at what might happening in the Guelph and Westmount-Ville Marie by-elections, which should be the most interesting to watch. The third by-election on Sept 8 is in the suburban Montréal riding of Saint-Lambert.
Bottom Line
Based on our estimates, we believe Bloc Québécois candidate Josée Beaudin begins the campaign with an advantage and could currently receive 9,800 to 10,920 votes (35-39%), Conservative candidate Patrick Clune between 7,560 and 8,680 (27-31%), Liberal candidate Roxane Stanners between 4,760 and 5,880 (17-21%), NDP candidate Richard Marois between 2,240 and 3,360 votes (8-12%), and Green candidate Diane Joubert in the 1,120 to 2,240 vote range (4-8%). Here’s how we arrive at these estimates (and they are just that — estimates).
Past Results
First, let’s recap what transpired in the January 2006 general election:
Party Candidate Votes % Riding
% Québec Spending BQ Maka Kotto 20,949 45.3% 42.1% $45,282 LIB Jean-Jacques Hermans 10,777 23.3% 20.7% $57,186 CON Patrick Clune 9,097 19.7% 24.6% $36,940 NDP Ronaldo Garcia 3,404 7.4% 7.5% $1,200 GRN Sonia Ziadé 1,819 3.9% 4.0% $0
As you can see, the Bloc and Liberals were slightly above average, the Conservatives slightly below average, and the NDP and Greens were average. However, since 2006, we have witnessed a strong growth in provincial ADQ support in the Montérégie region, which bodes well for the Conservatives. We therefore will estimate that the Conservatives will perform better than in 2006.
Current Trends
Now let’s see how party support has changed in Québec between January 2006 and now:
Party Jan 2006 Aug 2008 Difference Bloc Québécois 42.1% 37.4% -4.7% Conservative 24.6% 19.9% -4.7% Liberal 20.6% 23.3% +2.7% NDP 7.5% 11.0% +3.5% Green 4.0% 7.4% +3.4%
Estimated Base Support
There are a couple ways we might estimate (and we do mean estimate!) how general trending might impact support in Saint-Lambert: 1) take the baseline support from Saint-Lambert in 2006 and make adjustments for the changes in support, and 2) take the current province-wide support and adjust for the strength of Saint-Lambert relative to a party’s province-wide strength. In both cases, we need to then pro-rate them to 100%. These are really back-of-the-napkin calculations.
Using baseline strength (adjusted):
Bloc Québécois — 45.3 – 4.7 = 40.6 — 39.1% (pro-rated)
Conservative — 19.7 + 3.7 = 23.4 — 22.5%
Liberal — 23.3 – 1.7 = 21.6 — 20.8%
NDP — 7.4 + 3.5 = 10.9 — 10.5%
Green — 3.9 + 3.4 = 7.3 — 7.0%
Using current support/relative strength:
Bloc Québécois — 37.1 x 100% = 37.1% — 37.2% (pro-rated)
Conservative — 19.9 x 120% = 23.9% — 24.0%
Liberal — 23.3 x 95% = 22.1% — 22.2%
NDP — 11.0 x 90% = 9.9% — 9.9%
Green — 7.4 x 90% = 6.7% — 6.7%
Averaging the two methods, gives us:
Bloc Québécois — 38.2%
Conservative — 23.3%
Liberal — 21.5%
NDP — 10.2%
Green — 6.9%
Adjusting for Candidates
But, we must further adjust this because the Bloc incumbent is not re-running. Based on our modeling, we estimate a reduction of the Bloc vote by roughly 4 points. And based on second-choice preferences, these votes will likely be split, roughly 2.4 to the Conservatives, 1 to the NDP, and 0.6 to the Liberals, giving us roughly:
Bloc Québécois / 38.2 – 4 — 34.2%
Conservative / 23.3 + 2.4 — 25.7%
Liberal / 21.5 + 0.6 — 22.1%
NDP / 10.2 + 1 — 11.2%
Green / 6.9 + 0 — 6.9%
Adjusting for Campaign Spending
But we still haven’t adjusted for the fact that more financial resources are being put into the riding by the Conservatives and Bloc as compared to 2006. Last time the Liberals spent 74% to the maximum, the Bloc spent 60%, the Conservatives 48% and the NDP and Greens spent nothing. Expect the Conservatives and Bloc to run fully-funded campaigns, the Liberals about the same as 2006 and the NDP and Greens to run minimal campaigns. So we might expect the Bloc and Conservatives to improve their vote, which will likely force the Liberal vote down. Using our modeling, we estimate that accounting for campaign resources, we might see the following:
Bloc Québécois — 34.2 + 2.4 = 36.6%
Conservative — 25.7 + 3.7 = 29.4%
Liberal — 22.0 – 3.6 = 18.6%
NDP — 11.2 -1.5 = 9.7%
Green — 6.9 – 1.2 = 5.7%
Factoring Margin of Error and Turnout
Now if we take into account the margin of error in our current rolling average of polls in Québec of about +/- 2% and an estimated 28,000 voters (30% turnout), we might expect the following results:
Josée Beaudin, Bloc Québécois — 9,800 to 10,920 votes (35-39%)
Patrick Clune, Conservative — 7,560 to 8,680 votes (27-31%)
Roxane Stanners, Liberal — 4,760 to 5,880 votes (17-21%)
Richard Marois, NDP — 2,240 to 3,360 votes (8-12%)
Diane Joubert, Green — 1,120 to 2,240 votes (4-8%)
The above is purely an estimate based on deductive logic. What this suggests is that Bloc Québécois remain the favourite, but facing a decent challenge from the Conservatives. It will be interesting to see how this plays out.
Earlier, we took a look at what might happening in the Guelph by-election, which should be the most interesting to watch. But almost as interesting is Westmount-Ville Marie, a traditional Liberal strong-hold where star candidate Marc Garneau is facing a strong challenge.
Bottom Line
Based on our estimates, we believe Liberal candidate Marc Garneau begins the campaign with an advantage and could currently receive 9,600 to 10,800 votes (32-36%), NDP candidate Anne Lagacé Dowson could receive 6,900 to 8,100 votes (23-27%), Green candidate Claude William Genest and Conservative candidate Guy Dufort both at 4,500 to 5,700 (15-19%) and Bloc Québécois candidate Charles Larivée in the 1,800 to 3,000 vote range (6-10%). Here’s how we arrive at these estimates (and they are just that — estimates).
Past Results
First, let’s recap what transpired in the January 2006 general election:
Party Candidate Votes % Riding
% Québec Spending LIB Lucienne Robillard 18,884 45.7% 20.7% $70,313 CON Louise O’Sullivan 7,295 17.6% 24.6% $27,009 NDP Eric Steedman 6,356 15.4% 7.5% $13,082 BQ Sophie Frechette 5,191 12.6% 42.1% $9,770 GRN Julie Sabourin 3,451 8.4% 4.0% $122
As you can see, the Liberals more than doubled their province-wide share of the vote in the riding, as did the NDP and Greens. The Conservatives were about 30% below the province and the Bloc received less than a third of their province-wide average.
Current Trends
Now let’s see how party support has changed in Québec between January 2006 and now:
Party Jan 2006 Aug 2008 Difference Bloc Québécois 42.1% 37.4% -4.7% Conservative 24.6% 19.9% -4.7% Liberal 20.6% 23.3% +2.7% NDP 7.5% 11.0% +3.5% Green 4.0% 7.4% +3.4%
Estimated Base Support
There are a couple ways we might estimate (and we do mean estimate!) how general trending might impact support in Westmount: 1) take the baseline support from Westmount in 2006 and make adjustments for the changes in support, and 2) take the current province-wide support and adjust for the strength of Westmount relative to a party’s province-wide strength. In both cases, we need to then pro-rate them to 100%. These are really back-of-the-napkin calculations.
Using baseline strength:
Liberal — 45.7 + 2.7 = 48.4% (pro-rated)
NDP — 15.4 + 3.5 = 18.9%
Conservative — 17.6 – 4.7 = 12.9%
Green — 8.4 + 3.4 = 11.8%
Bloc Québécois — 12.6 – 4.7 = 7.9%
Using current support/relative strength:
Liberal — 23.3 x 200% = 46.6% — 43.0% (pro-rated)
NDP — 11.0 x 200% = 22.0% — 20.3%
Green — 7.4 x 200% = 14.8% — 13.7%
Conservative — 19.9 x 70% = 13.9% — 12.8%
Bloc Québécois — 37.1 x 30% = 11.1% — 10.2%
Averaging the two methods, gives us:
Liberal — 45.7%
NDP — 19.6%
Conservative — 12.9%
Green — 12.8%
Bloc Québécois — 9.1%
Adjusting for Candidates
But, we must further adjust this because the Liberal incumbent is not re-running. True Marc Garneau is a star candidate, but so to is Anne Lagacé Dowson, although probably not as much. Based on our modeling, we estimate a reduction of the Liberal vote by roughly 6 points. And based on second-choice preferences, these votes will likely be split, roughly 2.5 to the NDP, 2 the Conservatives, and 1.5 to the Greens, giving us roughly:
Liberal / 45.7 – 6.0 — 39.7%
NDP / 19.6 + 2.5 — 22.1%
Conservative / 12.9 + 2.0 — 14.9%
Green / 12.8 + 1.5 — 14.3%
Bloc Québécois / 9.1 + 0 — 9.1%
Adjusting for Campaign Spending
But we still haven’t adjusted for the fact that more financial resources are being put into the riding by the Conservatives, NDP and Greens as compared to 2006. Last time the Liberals spent 90% to the maximum. The Conservatives spent only 35% of the maximum. The NDP spent about 17% and the Greens spent nothing at all. I expect the NDP and Conservatives to run a fully-funded campaign and the Greens perhaps a 50% campaign (I would imagine the Bloc will run a minimal campaign). So we might expect the NDP, Greens and Conservatives to all improve their vote against the Liberals. Using our modeling, we estimate that accounting for campaign resources, we might see the following:
Liberal — 39.7 – 5.8 = 33.9%
NDP — 22.1 + 2.5 = 24.6%
Green — 14.3 + 3.0 = 17.3%
Conservative — 14.9 + 1.7 = 16.6%
Bloc Québécois — 9.1 – 1.6 = 7.5%
Factoring Margin of Error and Turnout
Now if we take into account the margin of error in our current rolling average of polls in Québec of about +/- 2% and an estimated 30,000 voters (30% turnout), we might expect the following results:
Marc Garneau, Liberal — 9,600 to 10,800 votes (32-36%)
Anne Lagacé Dowson, NDP — 6,900 to 8,100 votes (23-27%)
Claude William Genest, Green — 4,500 to 5,700 votes (15-19%)
Guy Dufort, Conservative — 4,500 to 5,700 votes (15-19%)
Charles Larivée, Bloc Québécois — 1,800 to 3,000 votes (6-10%)
The above is purely an estimate based on deductive logic. What this suggests is that Liberals remain the favourite, but facing a strong challenge from the NDP. It will be interesting to see how this plays out.
Updated: 14 Aug 2008
The most interesting by-election to watch on September 8 will be in Guelph, where four parties (Liberal, Conservative, NDP and Greens) are all running fully-funded campaigns and each is competitive. But who will win? Unlike general elections when past polls have shown that 80-85% of voters cast their ballot according as party votes (and only 15-20% because of the local candidate), we know that by-elections behave differently. They are much more candidate-focused since the government is not on the line. They can also be disproportionately influenced by a current “hot button” issue or a local issue. So the best we can do, in the absence of internal polls being leaked to us (hint, hint), is to use deductive logic.
Bottom Line
Based on our estimates, we believe Liberal candidate Frank Valeriote begins the campaign with a slight edge and could currently receive 9,800 to 11,200 votes (28-32%), Conservative candidate Gloria Kovach could receive 9,100 to 10,500 votes (26-30%), NDP candidate Tom King 7,000 to 8,400 votes (20-24%) and Green candidate Mike Nagy could receive in the 6,300 to 7,700 vote range (18-22%). Here’s how we arrive at these estimates (and they are just that — estimates).
Past Results
First, let’s recap what transpired in the January 2006 general election:
Party Candidate Votes % Riding
% Ontario Spending LIB Brenda Chamberlain 23,662 38.4% 39.9% $56,610 CON Brent Barr 18,342 29.8% 35.1% $82,313 NDP Phil Allt 13,566 22.0% 19.4% $30,298 GRN Mike Nagy 5,372 8.7% 4.7% $28,878
So, in both the 2006 federal and 2007 provincial elections, the Liberal candidate was roughly average relative to the province, the Conservative candidate ran noticeably weaker, the NDP candidate ran slightly above average. And the Green candidate ran considerably stronger, almost twice as strong as their average across the province.
Current Trends
Now let’s see how party support has changed in Ontario between January 2006 and now:
Party Jan 2006 Aug 2008 Difference Liberal 39.9% 39.2% -0.7% Conservative 35.1% 34.3% -0.8% NDP 19.4% 15.3% -4.1% Green 4.7% 10.4% +5.7%
Estimated Base Support
There are a couple ways we might estimate (and we do mean estimate!) how general trending might impact support in Guelph.: 1) take the baseline support from Guelph in 2006 and make adjustments for the changes in support, and 2) take the current province-wide support and adjust for the strength of Guelph relative to a party’s province-wide strength. In both cases, we need to then pro-rate them to 100%. These are really back-of-the-napkin calculations.
Using baseline strength:
Liberal / 38.4 – 0.7 = 37.7 — 38.1% (pro-rated)
Conservative / 29.8 – 0.8 = 29.0 — 29.3%
NDP / 22.0 – 4.1 = 17.9 — 18.1%
Green / 8.7 + 5.7 = 14.4 — 14.5%
Using current support/relative strength:
Liberal / 39.2 x 95% = 37.6 — 36.6% (pro-rated)
Conservative / 34.3 x 85% = 29.2 — 28.4%
NDP / 15.3 x 110% = 16.8 — 16.3%
Green / 10.4 x 185% = 19.2 — 18.7%
Averaging the two methods, gives us:
Liberal — 37.4% (pro-rated average)
Conservative — 28.8%
NDP — 17.2%
Green — 16.6%
Adjusting for No Incumbent
But, we must further adjust this because the Liberal incumbent is not re-running. Based on our modeling, we estimate this could reduce the Liberal vote by roughly 4 points. And based on second-choice preferences, these votes will likely be split between the other 3 parties, roughly 2 to the Conservatives, 1.2 to the NDP and 0.8 to the Greens, giving us roughly:
Liberal / 37.4 – 4.0 — 33.4%
Conservative / 28.8 + 2.0 — 30.8%
NDP / 17.2 + 1.2 — 18.4%
Green / 16.6 + 0.8 — 17.4%
Adjusting for Campaign Spending
But we still haven’t adjusted for the fact that all four are running fully-funded campaigns. Last time the Conservatives ran a fully-funded campaign. The Liberals spent only 70% of the maximum. The NDP spent about 36% and the Greens about 35%. Clearly, with fully-funded campaigns, we might expect the NDP and Greens to improve their vote. It’s likely the Liberals, given significantly more spending by the NDP and Greens, will likely hold steady despite slightly more spending. And given they already spent to the max in 2006, it’s likely the Conservatives will suffer with increases NDP and Green spending. Using our modeling, we estimate that accounting for all fully-funded campaigns, we might see the following:
Liberal — 33.4 – 3.0 = 30.4%
Conservative — 30.8 – 2.7 = 28.1%
NDP — 18.4 + 2.7 = 21.1%
Green — 17.4 + 3.0 = 20.4%
Factoring Margin of Error and Turnout
Now if we take into account the margin of error in our current rolling average of polls in Ontario of about +/- 2% and an estimated 35,000 voters (30% turnout), we might expect the following results:
Frank Valeriote, Liberal — 9,800 to 11,200 votes (28-32%)
Gloria Kovach, Conservative — 9,100 to 10,500 votes (26-30%)
Tom King, NDP — 7,000 to 8,400 votes (20-24%)
Mike Nagy, Green — 6,300 to 7,700 votes (18-22%)
The above is purely an estimate based on deductive logic. Clearly what this suggests is that all four parties are competitive. One thing is clear, it should be fun to watch.