Friday, November 1, 2013

Is Mulcair's QP performance paying off?

The problems regarding the PMO and Mike Duffy have certainly hurt the Conservative government, but the performance of Thomas Mulcair in Question Period has played a big part in the Prime Minister's discomfort. While Mulcair stole the show, Justin Trudeau reaped the benefits in the polls. Until now?

Considering the results of Ipsos-Reid's latest poll, as well as his improving approval ratings recorded by EKOS Research, it is a question worth asking.
Ipsos-Reid was in the field between the first and second explosive speeches made by Duffy in the Red Chamber, so the fallout from his last set of allegations (if there will be any) have yet to be fully recorded. Nevertheless, the poll has some interesting results.

(Ipsos-Reid also polled between Oct. 16-20, before the Senate scandal blew up again. The national results of that poll have been included in the lower bar in the chart above.)

The Liberals and New Democrats tied for the lead in the poll with 31% apiece (if you're wondering why I decided to put the Liberals first in the chart, it is because Ipsos had 263 people in their survey supporting the Liberals and 262 supporting the NDP - it is just that close!). Compared to their pre-Duffy polling, that represented a drop of two points for the Liberals and a gain of four points for the New Democrats. Those shifts are within the margin of error (of a probabilistic sample). The Conservatives dropped a single point to 30%, putting them in third place for the first time since, I believe, ever (at least for this current version of the party).

The Bloc Québécois had 6% while the Greens had just 2% support. 1% of Canadians supported another party and, by my estimate, 21% of those sampled were undecided.

It is perhaps more instructive, however, to look at Ipsos-Reid's previous poll from mid-September. Since that poll, both the Tories and Liberals dropped two points but the NDP picked-up five: that is outside the margin of error, suggesting that the New Democrats are really making gains.

This is certainly a great poll for the NDP, who have not led or been tied for the lead in any survey since October 2012. A word of caution, though, as Ipsos-Reid finds the NDP's supporters to be less committed than those of the Liberals and Conservatives. Two-thirds of their supporters were absolutely certain they would vote in an election (barring an emergency), compared to 58% for the NDP.

The source of the NDP's gains appears to be Ontario. The party tied with the Conservatives with 33% apiece in the province, a gain of eight points since September (the Liberals were at 32%). That increase was outside the margin of error, and the only shift since either the pre-Duffy or September poll to be outside the margin of error at the regional level anywhere. But with the exception of Atlantic Canada, the NDP did make modest gains throughout the country.

In Quebec, the NDP was up to 32%, putting them ahead of the Liberals, who were down to 30%. The Bloc Québécois fell to 25%, while the Conservatives were at 12%.

The New Democrats also led in British Columbia with 36%, followed by the Conservatives at 29% and the Liberals at 28%. The Liberals led only in Atlantic Canada, with 55% to 26% for the Tories and just 17% for the NDP.

The Conservatives were ahead in Alberta with 49% to 26% for the NDP and 19% for the Liberals, while they were in front with 46% to 29% for the NDP and 24% for the Liberals in the Prairies. It is perhaps worth noting that in both Alberta and the Prairies, second place flipped from the Liberals to the NDP between the Oct. 16-20 and Oct. 25-28 polls. The shifts were within the margin of error, however.

It is also worth noting generally that this Ipsos-Reid poll stands out for having the Liberals low and the NDP high, particularly in Ontario. We will have to wait and see whether other polls will confirm that the New Democrats are making gains.

These levels of support would yield an interesting House of Commons. The Conservatives would likely win some 121 seats, with the NDP and Liberals tying with 104 apiece (if you're now wondering why I have the NDP in second in the chart, it is due to the convention that parties maintain their previous role occupied in the House in the case of a tie). The Bloc would take eight seats and the Greens would win one.

The Conservatives benefit from the divided opposition in Ontario, winning the plurality of seats in the province. They also put up good numbers in the West, but lose quite a few seats in B.C.

The New Democrats win the majority of seats in Quebec and manage a strong performance in Ontario and British Columbia, along with a smattering of seats on the Prairies and in Atlantic Canada. The Liberals win most of their seats east of Manitoba.

It is hard to imagine that the Conservatives would be able to govern very long, if at all, with this sort of result in the House of Commons. But it is also hard to imagine what the Liberals and New Democrats would do. Govern in tandem? With a tie in both popular vote and seats, who would take the lead in such a coalition? Would Mulcair or Trudeau be Prime Minister? The most likely result, in the end, might be another election to settle the matter since there is no other simple solution.

The EKOS poll

The poll by EKOS Research, done over most of the same days as Ipsos-Reid, had a very different result: 37% for the Liberals, 26% for the Conservatives, and 25% for the New Democrats. The numbers, even at the regional level, were broadly similar to their previous poll from earlier in October, which I analysed here. Interestingly, EKOS shows that Mulcair's approval ratings have soared to almost 50%, a big change from previous polls, both their own and those by other firms.

But there is something that is important to note with the EKOS Research poll. Brought upon by questions asked by Abacus Data's David Coletto, Frank Graves of EKOS wrote up an explanation of why his latest poll was heavily weighted towards university graduates, to the tune of almost three times the number it should have been compared to the census. I have brought up the problems related to over-sampling one group before (most recently here), but the potential issue with the EKOS survey is that the poll was not weighted by education at all, meaning that the poll is skewed towards those with a higher level of education (rather than just having the risk of the lower-sampled groups having their errors magnified by inflating their weight).

Coletto found that if the sample was weighted according to the census results for education levels, the results would be dramatically different: 33% for the Conservatives, 32% for the Liberals, and 24% for the NDP. This would bring the poll's results much more into line with what Ipsos-Reid, Abacus Data, and Harris-Decima have recently recorded. The same would happen with EKOS's previous poll in September.

Graves's response to Coletto's concerns is laudable. It is hard to imagine many other pollsters being so open with a methodological issue like this, and we should applaud this degree of transparency. It is far too rare in the industry.

The main element of Graves's response is that respondents are known to inflate their level of education - a high-school dropout becomes a high-school graduate, a college degree becomes a university degree, a few years at Trent becomes a bachelor's, etc. That makes this more difficult to compare a survey's education-level sample to the census than, say, gender or age. Graves suggests re-weighting his sample to a different proportion for education, with 45% being university educated instead of the 22% from the census or the 62% from his poll. The results change accordingly: 35% for the Liberals, 29% for the Conservatives, and 25% for the NDP.

I'm not sure I quite agree with Graves's re-weighting approach, as it would seem to make a lot more sense to bring the numbers much closer than that to the census. But I do understand his concerns with the difficulty in getting accurate responses (if you weighed a poll by ethnicity, what do you do with all those people who impossibly say their ethnicity is Canadian?). On the flip side of the coin, though, if the sample itself cannot be trusted, why report the results for that particular category? Are university graduates more likely to support the Liberals, or only people who falsely claim to be university graduates? How useful is that information?

If the results were broadly similar across the different educational levels, this would not be an issue at all. But since they are significantly different, and have been for some time since Trudeau's surge, this is an important issue. I would suggest that EKOS weigh their polls more with a nod towards educational levels than they have in the past, since it seems to be a leading indicator of how someone will vote: 42% of self-proclaimed university graduates support the Liberals, compared to 28%-29% of the rest of Canadians in the EKOS poll. The NDP's support was mostly uniform, while the Conservatives get 36%-37% support among lower educated voters, and 21% among university graduates.

Another point Graves's brings up, which I do agree with, is that pollsters need to draw a line of what to weigh a sample by. Any number of factors could be chosen - should you weigh a poll by beer preference according to market share? But education is a rather basic measuring stick and, perhaps more relevantly, it is a category that EKOS itself reports.

On that score, the sort of concerns brought up by Coletto and on this site in the past give pollsters good reason to limit the amount of information that they release with their polls. Why run the risk of some pesky nerd poking holes in your work? There is, of course, a big difference between the right thing to do and the easy thing to do. EKOS is doing the right thing, and the openness with which Graves approaches methodological issues gives me more confidence in his work than it would otherwise. Other pollsters, including some of the most important ones in Canada, are asking us to take their numbers on faith. We should instead be expecting and demanding the degree of transparency we get from firms like Abacus Data, Ipsos-Reid, and EKOS Research.

26 comments:

  1. Éric,

    Some of us thought Mulcair was no Jack Layton, even though I like him. Well, he is clearly on the right track and it's translating into increased party support. Liberals are still viewed in many minds as the natural alternative but people seem to be unsure whether to move strategically to Justin. Either way, the momentum is with the opposition parties. Harper is not out of it yet but the trend is not his friend. Boy, will he ever have his work cut out for him if the CPC is to recover. That's a tall order but not totally impossible. However, I won't bet any money on the Conservatives bouncing back.

    ReplyDelete
  2. Éric, I have to take issue with you describing a Canadian ethnicity as "impossible". ("...those people who impossibly say their ethnicity is Canadian...") I have to believe that it's just a possible as "Latin American", "British", "Belgian", "Pakistani", "Taiwanese", etc. All of those were created by defined borders that didn't particularly reflect cultural or genetic groupings, or are heavily affected by immigration. I'm not actually saying that those folks *are* or *aren't* Canadian "ethnically", only that all ethnicity is, to some degree, a choice of definition. My family has been in North America for 15 generations, and I have no family ties to Ireland or Germany, where many of my ancestors originated. Does that mean I have NO ethnicity?

    In general, I think that ethnicity is a silly way to categorise the world (as opposed to "cultural groups", which are far more useful, if much harder to count), but if it is going to be used, then "Canadian" is just as possible a label as any of the others.

    ReplyDelete
    Replies
    1. I disagree. My ethnicity is French and Irish, and that is just what it is. It doesn't matter that my ancestors have been here since the 17th century. It isn't a made-up construct, it is in our DNA: http://en.wikipedia.org/wiki/Genealogical_DNA_test

      Delete
    2. What part of France? A French national identity only really fully formed in the 18th century. Are you sure your ancestors wouldn't have considered themselves as Saintongeais or Norman perhaps?

      FYI, those DNA tests only tell you what areas a certain haplogroup is most common in. They're very widely dispersed, and gradually increase and decrease as a gradient, not some hard cut-off. The most common Y-chromosome haplogroup in both Ireland and France is the R1b haplogroup. It's still found at detectable frequencies as far east as western China, and is the most common haplogroup in some parts of Cameroon even. If you delve deeper you might be able to get some idea of where your direct male-line and female-line ancestors were living at a given period of time, but that tells you absolutely nothing about all of your other ancestors, from whom a majority of your DNA came. You can't tell if someone is Irish or French from their DNA at all.

      If you took someone at random from Istanbul today, odds are they'd self-identify as Turks and speak Turkish. Go back 1000 years, and most of the ancestors of this same person would identify as Greek or Roman and speak Greek. Go back 1000 years before that their ancestors would speak Greek or Thracian. Another 1000 years prior, they may have been speaking Phoenician. Yet genetically, this is the same continuum of people.

      Delete
    3. Put it this way - you consider Irish an ethnicity with a basis in genetics. The dominant genetic marker for men in Ireland is the R1b Y chromosome haplogroup. The best guess right now is that the R1b haplogroup arrived from northern Spain with the expansion of agriculture to Ireland around 4000 BC.

      The Celtic languages that became Irish didn't arrive in Ireland 3400 years later with the Celtic expansion from central Europe during the Iron Age. Those languages, along with other Indo-European languages (ie English, French, German, Greek, Persian, Urdu, Punjabi, Hindi), can trace their origins back to the domestication of the horse in the southern Ukraine around 3,700 BC.

      So, could you explain to me how Irish ethnicity can be based in genetics that arose 3400 prior to any Celtic language ever being spoken in Ireland?

      Race and ethnicity are cultural construct. Nothing more and nothing less.

      Delete
    4. Which language to we identify ourselves by too? Is it the one that our ancestors spoke the longest? If so, you and I are more correctly Basques than Irish-French or German-Scotch.

      Delete
    5. I imagine my ancestors would have considered themselves Norman, but I'm not entirely sure. I just finished reading a book on Champlain, and while he definitely considered himself a Saintongeais, there was also a strong streak of 'Frenchness' in the language of the time. Your point is taken, but that seems to be more about national identity than ethnicity, which are two different things.

      I clearly don't know about this topic as much as you do, so I will yield the floor. But I don't know if ethnicity is something considered to be something passed down over many millenia. While I do consider myself to share some genetic material with the Gauls of the Roman times, I don't really consider myself Gallic. Stretched to its logical extension, we're all sub-Saharan Africans.

      It is an interesting topic, to be sure, and I'm not sure how uniform it is. Think of eastern Europeans and the various borders that came and went over the centuries and the movements of people, but compare that then to Icelanders who are considered one of the most genetically uniform groups in the world, or other generally insular cultural groups like, say, French Quebecers, who share a very shallow genetic pool.

      Delete
    6. Certainly First Nation peoples should be considered "ethnically" Canadian. Although I suspect many would say we're Cree not Canadian or Mohawk or Iroquois etc... Ethnicity is both a social and ancestral group of people. An immigrant could be Canadian so long as he identified with the culture's social traits or characteristics. Conversely a person may not have social attachments but, be a member of a certain ethnic group by ancestry. For example a Quebecois could be Quebecois, Gallic but, not French. Or an Englishman could be genetically Scandinavian yet still be English.

      In any case I think these definitions change over time so I see no reason to not include a Canadian ethnicity.

      Delete
    7. I'd argue even that ancestry aspect is cultural, in the sense that it is one's perception of their ancestry that is relevant. Whether I'm part of the R1a or R1b haplogroup doesn't really affect my daily life, and without testing I wouldn't be aware of which group I'm a part of. Without being aware of some distant horse warrior ancestor from the Russian steppe, I'd not be likely to identify with such a community.

      I do think you can be part of multiple ethnic groups though. I consider myself first and foremost Canadian in ethnicity, but I also consider myself Scottish and German. One of my best friends considers himself Punjabi and Canadian. He is no less Canadian than I.

      Is there some biological/genetic correlations within ethnic groups? Sure, but that shouldn't be surprising. People living in the same area tend to both have kids with one another and adopt one another's language and customs.

      Delete
    8. You're correct he is no less Canadian than you - because it is a nationality! He can't consider himself ethnically German, though, and you can't consider yourself ethnically Punjabi, no matter what either of you would do...

      And I disagree strongly that ancestry is about perception. Our ancestors have a direct influence on our lives, be it via genetics or upbringing. My father was influenced by the way his father raised him, and his father was influenced by the way his father raised him, and so on and so on. Whether I perceive it or am aware of it is irrelevant.

      Delete
    9. ...Unless you were adopted! Then you get a combination of familial and genetic heritage. ...and that's what I (and others) are saying, I think - that an "ethnicity" may be, in fact, very closely correlated to a genetic match (say, Iceland), or have nothing to do with it (The number of different genetic backgrounds that make up Cuba, for example, or Lebanon).

      I don't think that the argument is so much about how legitimate "Canadian" is as an ethnicity as it is about how arbitrary many of the "ethnic" labels are, hence bringing "Canadian" into the fold as "just as arbitrary".

      "Lebanese" is commonly considered an ethnicity in Canada, broadly unknown in Lebanon, where one's ethnic, religious and cultural makeup are very complex topic, and the only people who claim to be "Lebanese and nothing but" are usually people trying to make a poltical point about national unity (not unlike many "Canadians").

      "German" - made up of a gajillion smaller countries, but sharing a common language. Wait, but so do the "Austrians". So - can they be the same ethnicity?

      "Jewish" - I think that many people consider themselves ethnically "Jewish", particularly if they lived in countries where they experienced exclusion and ghettoisation for generations, However, that's clearly not the same thing as "Israeli", because many have never lived in Israel. On the other hand, picture a Jew who left Berlin before WW2 - he/she isn't "ethnically German" by any reasonable definition, but then - why should I be?

      So - like an adopted child, a genetic heritage that gets mixed in with others ends up with features of all of them - and if that mixture starts to look and act the same as the other nearby mixtures, even though they may have had initially-different ingredients then eventually that set deserves a name, and it's legitimate to use it.

      There aren't many people around calling themselves "part Carthiginian", but that was surely a reasonable ethnicity at one point in history. Times change, and though it probably once wasn't, "Canadian" is now, just as good as the other labels.

      Delete
    10. For most of history Austrians and Germans were considered the same ethnicity-German. The Holy Roman Empire was composed of hundreds of sovereign princely states but, not nations. The Emperor's subsidiary title was "King in Germany" much like the Kaiser's title was "German Emperor". These titles are deliberately vague and do not refer to a specific territory or jurisdiction.

      Delete
  3. Hi Éric; given how the post-mortem of the recent surprise elections in BC and Alberta focused on adjusting for turnout, and your experiment with your simple turnout model by age, I'm surprised you didn't acknowledge Graves' point that weighting by self-reported education "(could) begin to over-represent those who will not show up at a polling booth (those without post-secondary education are consistently less likely to vote on Election Day)".

    Admittedly, even in the Ekos brief it was a throw-away line.

    While turnout will continue to be difficult to estimate, I personally think that it remains the elephant in the room for political polling and electoral projections. Education could be as significant an effect as age or income.

    ReplyDelete
    Replies
    1. There is no problem with using this sort of consideration in determining likely-voter support, but this is not what the EKOS poll was about.

      Delete
  4. It's fine to say Mulcair is doing great and he is in Question Period. But in an election it's going to be a different scene.

    Mulcair's targeted lawyer approach works really well in the current Commons situation but not on the hustings !!

    I think Justin will do better on the hustings than Mulcair or Harper because of personality differences !!

    ReplyDelete
    Replies
    1. Mulcair has a good chance of dominating the leaders' debates, IMO. But of course, that's not what decides the election.

      Dom

      Delete
  5. Éric,

    As an ecologist who regularly tackles various statistical analyses I'm a little bit embarrassed to request this, but sometime why don't you do a whole piece on exactly how the infamous "margin of error" should be interpreted in the context of opinion polls?

    My understanding was that the margin of error reported by pollsters is a sort of upper limit which is only actually reached when one of the categories in the survey (e.g. support for a particular party) approaches 50%, but your statement that NDP support rising 4 points from 27% to 31% between two polls with margins of error of 3.4 pp is within the margin of error has me doubting myself. My assumption is that you consider that NDP support could've actually been as high as 30.4% in the first poll and as low as 27.6% in the new poll given their respective MoE's, and therefore it would've taken an increase from 27 to 34% for you to consider it statistically significant?

    In any case, I still think a general article on the interpretation of the MoE would be of interest to your readership. Cheers,

    Dom

    ReplyDelete
    Replies
    1. There is a formula for calculating the margins of error for changes in support of a party from one poll to the next, and it depends on the sample size and the level of support. I plug the numbers into the formula, and that tells me whether a shift is statistically significant or not.

      It might be worthwhile to go into it in more detail at some point, but I'm intrigued by the statement Nate Silver once made on his blog that the margin of error is not nearly as important as it is made out to be.

      Delete
    2. I'd agree with Nate Silver on that. Lay readers of statistics tend to consider that anything within the margin of error is equally probable, and anything outside the MOE is out of the question. In fact, the MOE is an arbitrary cutoff. The conventional MOE is for 95% confidence, meaning that the results are within the margin 95% of the time -- and approximately 1 out of every 20 polls you read is actually outside that margin!

      Delete
    3. Well in Canada at least, systemic bias seems to be substantially larger than statistical noise.

      Delete
  6. Evan
    I think Tom Mulcair is doing a fine job. I have met him 3 times and have been impressed. I think he is doing well on the hustings as the polls are indicating.

    ReplyDelete
    Replies
    1. The Hustings only appear at election time. The hustings are the traditional place of assembly or courts and or a raised platform where electors cast their vote. Today the word hustings refers to an election specifically and strangely it is one of the few English words that is time specific. Without an election a politician is simply glad handling not technically "on the hustings".

      Delete
  7. Imagine the fascinatingly unfair situation of the third place party getting the plurality of seats. Neither of the two more popular parties would want to form a coalition with each other without their own leader being prime minister, since (a) the Ontario experience of the 80s and the UK and German experience of the 2000's show that the senior coalition partner tends to outshine the junior partner, (b) both parties would expect the governing party to be weak and unstable, forcing another election relatively soon (and thus a chance at winning rather than coming oh-so-close). So, due purely to the Prisoner's Dilemma Paradox, the third place party could end up governing for quite a while, which might convince even the most placid Canadian that the single-member-plurality electoral system is ridiculous.

    ReplyDelete
  8. Hi GI,

    I agree with your analysis although how unfair the situation may be is an individual perspective.

    In the situation above I think you're correct; the dilemma of who from the first or second party becomes PM may Scotch (the ethnicity not the drink) either of them becoming government. If Harper resigned and the GG had to summon a particular leader, Trudeau or Mulcair, a number of criteria may enter the GG decision: popular vote, representation in all or most regions, ability to make the current Parliament work, ability to introduce legislation in the Senate, constitutional conventions etc...Whoever he asks may decline appointment or both may in favour of a dissolution.





    ReplyDelete
    Replies
    1. bede dunelm,

      Call me old fashioned but in my book all that counts is whether the incoming PM (Prime Minister Designate) can command the confidence of the House.

      Delete
    2. The Monarch or Governor must have the best interests of the country, monarchy and democracy at heart when making a decision of this magnitude.

      In a practical sense most prime ministers or premiers come into office without having tested the will of the House. We tend to over look this fact because most have majority governments and confidence is assured.

      In the situation above if Harper resigned either of his own accord or because of loss of confidence the GG would need to pick a replacement. Once appointed the new PM will test the will of the House. The closeness of the results seem to complicate matters as a prima facie reading indicates no party alone has the ability to form a workable majority. Although by convention the Official Opposition (NDP) would be offered the first opportunity to form government given the near even results I think the ability to produce a workable Parliament would be foremost on the GG's mind and that may favour the Liberals. The outgoing PM also has the ability to advise on his replacement although it is up to the discretion of the Monarch or Governor whether to act on it or not.

      Delete

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.