Alright, it's almost over! One week exactly and Elections BC will stop accepting ballots and will start counting them. Turnout is now at 37% (received ballots; it's 31.1% of processed ballots). So the good news is that the final turnout will be decent. More importantly, by being above the 40% mark, it ensures the Liberals leader Andrew Wilkinson won't be able to call the result illegitimate. Now the questions is really: who will win?

The turnout is actually still ahead of the one for the HST referendum:


Although we are closer to the deadline than we were for the HST vote. If we instead look at the number of days remaining, we are slightly behind:

It seems a finish around 45% is likely and this is much higher than I'd have thought.

Here below is the current turnout by riding. Remember that some regions received their ballots later and/or were more affected by Canada Post's strike.





So, who is likely winning?

Throughout this campaign, I tried to estimate the number of votes cast for the YES and NO side. Ever since I switched to using the turnout by age provided by Elections BC, I've had the NO side ahead but with a decreasing lead. It has been quite stable the last few days.



Remember that these are estimates. They are subjects to margins of error, in particular from the polls they are based on (technically I'm using turnout, demographic and polling data to provide these estimates).

So first, let's see how sensitives these estimates are to a change of specifications.

1. What if we apply some margins of error to the polling averages

I'm using the four published polls during this campaign (from Angus-Reid, Mainstreet, Insights West and Research & Co.; I'm not using my private Google Survey even though the results were very similar, mostly because I don't have as much details regarding the location of the respondents).

But those are polls, they have margins of error even when we aggregate them. The issue here is that I'm not sure how accurate these polls are. For federal or provincial elections, I have done the research and know the empirical margins of error. For referendums however, we don't have many data points. In general polls have done quite well for the 2009, 2011 or 2015 transit plebiscite, but it doesn't mean it'll always be the case. The number of undecided this year is much higher as well.

These polls have also been remarkably similar. So let's use the standard error of these four polls and create an interval by adding or subtracting 1.96 times this standard error. That gives us margins of error of plus or minus 1.5% (and more by age or regional breakdown but let's ignore this for now; Errors are likely to be correlated anyway. For instance if the polls underestimated the YES in the Lower Mainland, they'd have done the same on the Island).

Ok, so let's re-run the estimates above but using either the upper or lower bounds for the support for PR.

% for PR with upper bound: 50.4%
% for PR with lower bound: 47.4%

So the YES can win but the interval shows it's not the most likely outcome.


2. How important is the age turnout?

These estimates work this way: using the turnout by age of 2017 (thus telling me the percentage of voters aged 18-34 in each riding for instance) and the region, I look at the number of votes in one riding and "assign" them to the YES or NO based on the demographic data (the age thingy) as well as the levels of support for this age group in the polls.

The issue here is that around 50% of voters in 2017 were actually above 55 (this is a crazy stat as the 55+ do not represent remotely close to 50% of the population). Given that polls show the 55+ want to keep FPTP in majority (63% exactly), this gives the NO side a huge boost.

But here's the thing: polls always weigh their samples based on census data, not turnout one. At first it seems like a perfect recipe to be wrong (since you'll underestimate the impact of the 55+ and over estimate the 18-34 for instance) but in practice it has been relatively fine. Look at the polls for the BC election last year. They did a fairly good job. If you re-weight the polls with the age-turnout data, you usually get a higher number for the Liberals than what the polls showed. This is typical in pretty much every single election where one party does better among the 55+. But again, in the real life, we don't observe such parties to be systematically underestimated.

What I'm saying here is that my method above might actually be quite harsh against the YES side.

There is also the issue that regressions (see below or previous blog posts) have shown a clear trend: the 18-34 are voting more than in 2017 while the 35-54 are voting a lot less (the 55+ were voting more at first but that's not the case anymore). So there as well I might be underestimating the YES side (note: my method should technically account at least partially for this since more votes are coming from ridings with more voters aged 18-34, but I'm not making any additional adjustments such as increasing the share of the 18-35 at the riding level).

Finally, I would technically need the polling data broken down by age AND region (like what is the level of support for PR for the 18-34 in the Lower Mainland?). Polls don't provide me with this. They provide the average by age OR the region. My method above basically runs the estimation using each and does and does an average.

I contacted the pollsters to try to get the data I want (because they have it), we'll see if they nicely oblige. In the meantime, we can try one thing: let's assume the two (age and region) are independent from each other. So if we compare the levels of support for the 35-54 to the 18-34 in the Lower Mainland, we should get a proportionally similar result as comparing the 35-54 to the 18-34 on the Island for instance. This assumption will amplify the differences. For instance the 18-34 (more in favour in general) living on the Island (generally more pro-PR) will therefore be really, really pro-PR. On the other hand the 55+ in the Interior will be incredibly against it. That might actually be a good thing as I suspect that the two campaigns are motivating the more "extreme" voters (the ones really in favour of PR and those really against it).

If I do this instead, I get the following graph:



Same trends but the YES is now ahead. Interesting. The assumption of independence is really strong (and likely invalid) but at the same time, is it worse than doing both separately and averaging? Not sure. Please somebody smarter than me could let me know.


3. What about new voters?

8487 people registered to vote by the deadline of Elections BC. That's not a large number (it's 0.26% of the registered voters). However, it's reasonable to assume that a majority of these voters did so to vote for PR. Why would people who weren't registered decide to go and vote to keep a system they weren't even using? I'm sure there are some people like this (maybe to protect the province from all the Nazi parties that PR will inevitably brings if you listen to some people lol) but the majority of these 8487 should go into the YES side.

If we assume that 75% of these new voters will pick the YES side, then my estimated percentage for the YES increases by a little bit less than 0.2%. Minor? Yes, but that could well be the difference at the end.


4. What about using the votes for each party in 2017?

Maybe instead of trying to mix age and regional data, the best and easiest method is to take the 2017 results by ridings and use those as benchmark. So a riding with a majority of votes for the BC Liberals will vote against PR. This should capture both the age and regional effects.

Unfortunately for us, most polls did not provide the levels of support for each side by the political choice in 2017. Only Research Co. and Angus-Reid did so and we can see that the BC Liberals voters really, really hate PR (81% in favour of keeping the current system) while NDP and Green are mostly in favour (around 70-75%).

So now my calculations are as follow: imagine a riding voted at 40% for the Liberals. Therefore, 40% of the votes for the current referendum will be assumed to come from the BC Liberals voters and 81% of them will vote NO. The regressions below show that the 2017 turnout is a strong predictor of the current turnout, so this method might actually be quite valid.

If I use this, I get the following graph (yes I know this post has the same graph many times but that's the point of this article):



One potential problem with this method is that it doesn't account for how the other voters will vote this time (in particular the BC Conservatives). Still, we see similar patterns but here again the YES side seems narrowly ahead. The trend for PR is quite steep but the scale is misleading as it shows results really, really close to 50%. Interestingly this is the method that gives us the closest race so far and the smallest absolute variations (it stays between 50.8 and 49.2%).


5. Regressions

What the polls have shown is that people in BC are mostly split 50-50. So using the turnout by riding so far, we can try to find patterns (and stop trying to actually estimate and count votes). Here are the results with the latest numbers:


If you've been reading my blog during this referendum, you already know that the same regressions were initially showing a positive coefficients for the share of 55+. Over time it has switched from a positive and significant effect to a negative but not statistically significant one. In the less simple model, we are actually getting close to getting statistical significance.

As for the regions, the Lower Mainland and the Island are above once we adjust for the number of days they had to vote (which is, again, still a determinant of the turnout even this late into the referendum, thus showing once again that Elections BC was right to extend the deadline).

The major impact of the % of people with English as mother tongue is just confirming to us that the Vancouver suburbs of Richmond And Surrey aren't very interested in this process so far.

If I had used the percentage for each party (instead of looking at whether the riding elected a Liberal or NDP MLA last year), then the share of 55+ is negative and significant. And in this case the higher the % for the Green, the higher the turnout (significant at 10%) while it's the opposite for the % for the NDP (thus again showing that this party isn't getting its vote out as much, although this could be partially explained by Surrey).


Conclusion

It still a very close race but the trend remains favourable to the PR side. Margins of error make predicting this referendum almost impossible but there are reasons to be optimistic if you want to change the electoral system.

The regressions show that the 18-34 are voting more than they have in the past while it isn't the case for the 55+, at least using aggregate data (so subject to a possible ecological fallacy - Google it if you want to know more). Regionally, the effect of when the ballots were received (and/or of the postal strikes) are decreasing and we gradually see the Lower Mainland catching up the Interior while Vancouver Island is now firmly ahead. Both trends should be positive for PR.

On the other hand, ridings with a BC Liberals MLAs continue to vote more (we see it with Vancouver Quilchena for instance) and this is good news for the current system. I heard that the BC NDP was now making phone calls to motivate their voters to get out and mail their ballots. Hopefully this isn't too little too late.

Personally I'm becoming cautiously optimistic for the YES side but I'll wait to see the final turnout to really make a call.
Simple, interactive map of the turnout (processed ballots) as of November 30th 2018.

Four categories, two below the average and two above.

That is all. Hope you enjoy.

Interesting new turnout data published by Elections BC yesterday. Let's take a look.

Summary for busy people

Using all the information available so far (polls, turnout by riding, data about the turnout by age in 2017), here is my best estimate of the current situation in the referendum on electoral reform in BC.




The NO side (keeping FPTP) is still ahead but the trend is favourable to the YES side (switching to proportional representation). The gap was 4.8% at the beginning and is now only 2.3%. Over the last few days alone, the YES side has progressed by an estimated 1 point.

How do I estimate these votes? Knowing the number of votes from each riding and knowing the age profile and regional location of these ridings, I estimate how any of these votes are likely a YES or a NO. Without going into the details, a riding in the interior with a high share of voters over 55 is likely to vote in majority to keep the current system. The exact calculations involved the polling averages and the turnout by age of 2017.

If you are among the dozen of people who actually read my blog on this topic, you should remember that my previous post was saying the YES might have been ahead already. So what happened? Well I was using the census data to get an age distribution by riding. But Elections BC has something better: the turnout by age by riding in 2017 (and before). This is a more accurate measure of who is potentially voting, but this is also a measure that hurts the PR side because older voters represent a disproportionate share of the voters (see below). So it isn't that I changed my mind or trying to generate clicks on my site, it really is just an updated version of what I was doing. The good news is, I believe, that my model will now remain the same. If I were to use the census age profile of my last post, then PR would be ahead 50.6% to 49.4%.


1. Current turnout

Based on the data published by Elections BC on Friday, the turnout is already 30%, although it falls to 18.9% if we only count the ballots that have already been processed. The roughly 12-points gap seems to be fairly consistent (i.e: both measures increase at a similar rate). Extrapolating the trend for processed ballots until the 30th and assuming the gap remains the same, it seems the turnout could end up close (or even above) the 40% mark (which, incidentally, is the arbitrary threshold set by the BC Liberal leader Andrew Wilkinson to recognize this referendum as "legitimate"). This is higher than I was predicting at first. One reason is because my early predictions were based on the turnout among processed ballots (Elections BC only mentioned they had way more ballots than the ones processed relatively recently). It's actually not impossible that we end up with a higher turnout than for the HST referendum!

Using Google Search as a measure of the interest for this referendum, we can see a declining search activity since the peaked around the debate (November 7th). So I'm not sure we should expect a late surge in turnout, but we never know. This whole analysis is also made more complicated because of the possible impact of the rotating strikes of Canada Post.

Overall though, the turnout will go from decent (low 30s) to quite good (40% and more). For a referendum coming in November, after a municipal election and regarding a topic that isn't the most interesting for many people, I consider this a small victory. Fears of ridiculously low turnout that would make this referendum invalid can be set aside it seems.

The current turnout is still heavily influenced by when the riding was scheduled to received the ballots, although the November 2nd group is clearly catching up to the 23 ridings that got their ballots early.



As explained in a previous post, the scheduled date from Elections BC isn't perfect and we know for a fact that ridings on the same schedule didn't received their ballots at the same time. This is why I prefer using my measure of "days where the turnout has been over 1%" as my measure of how many true days each riding has had to vote.


There is still a strong relationship but it's converging for sure. There shouldn't be any relationship by the end of this process (if there is, then this could be problematic as it'd mean the ridings that got lucky and received their ballots earlier voted more).


2. Analysis of the preliminary turnout

First, let's look which region is voting more (the Lower Mainland, Vancouver Island or the Interior). As of Friday, here were the turnouts (screened ballots):

Lower Mainland: 17.2%
Island: 21.1%
Interior: 19.9%

As comparisons, during the 2017 BC elections, these rates were, respectively, 59.7%, 66.4% and 60.2%. Also, and maybe more importantly for this referendum, the last few days have seen a dramatic increase in the Lower Mainland. It seems the many ridings that got their ballots later started voting (or Elections BC started counting them). Just 3 days ago the turnouts were, again respectively, 8.8%, 14% and 15%. The regions that are known to be more pro PR (Lower Mainland and the Island -where the BC Green are the strongest) have caught up with the interior (and actually surpassed it for the Island).

This trend is the big reason for the trend in favour of the YES side in the graph above. If it were to continue until Elections BC is done receiving and counting the ballots, I think the chances of a victory for the YES would become quite good.

The region isn't the only important factor. Polls have all shown that age is a crucial determinant of the level of support for proportional representation.

Looking at the 2017 election, we can see how much the 55+ voted more than the 18-34. Here is the turnout by age group for that election:

18-34: 49.3%
35-54: 57.7%
55+: 69.9%

In average, the 18-34 represented only 18% of the voters while the 55+ accounted for 49.8% of the ballots cast. So there is no question that the fact FPTP is the most popular option among the 55+ crowd is making switching to proportional representation difficult. Turnout in 2013 were also fairly similar.


3. Regression

We obviously don't know who is voting in each riding, we just know the total number of votes so far. Still, we can try to see some patterns. But for this, we need a regression. If you don't know anything about statistics, see a regression as correlations measured while holding other factors constant. So since ridings in the interior also have more older voters, it's hard to disentangle which variable is affecting what. A regression solves this problem. The significance level is simply a statistical measure of how confident we are of the results. If there is no * next to the coefficients, it means it isn't significant and we consider that there isn't a relationship between the turnout and this variable. The more stars, the more confident we are (* for 10% uncertainty, ** for 5% and *** for less than 1%).

What I want to know now is if, after controlling for the number of days each riding had to vote and the region, the current turnout is still higher in ridings with more voters aged 55 and over.

The table below shows you the results:


As mentioned above, I'm now using the actual turnout by age to determine the age composition of each riding. Thus instead of using the proportions of people aged 18 to 34, I'm using the share of voters that were 18-34 in 2017. I'm including the turnout in 2017 as a control. I really want to measure differences between this referendum and the previous election. Why? Because if we were to observe a significant different (for instance 55+ voting much more), then my vote estimates of above would likely be incorrect.

What we see is that ridings with a bigger share of 55+ are NOT voting more anymore. At least if compared with the 18-34 group (this is really how you should interpret the coefficients). It seems this referendum is motivating two groups to vote: young people (who want PR) and older folks (who want to keep the current system). The 35-54 (who are the most 50-50 on PR based on polls) seem less interested.

Just a technical note here: I'm using aggregate data (at the riding level) and therefore my regressions are subject to the typical ecological fallacy. What is this? Let's use an example. In the US, traditionally, richer people vote Republican (that might have changed with Trump but let's ignore him). At the same time rich states (NY, California) vote Democrat. So the correlation at the aggregate level doesn't match with the one at the individual one. In BC, we observe a similar effect where ridings that vote NDP have a higher proportion of Asian people but polls have shown that Asian people are more likely to vote for the BC Liberals. What does it mean for the current regression? Well hard to tell, but you should at least keep in mind that I'm not actually observing the age of the current voters, only that ridings with more older and younger voters are voting more. For instance, imagine a scenario where this referendum is motivating young people in ridings with a higher share of older voters, then the higher turnout from those ridings would actually be favourable to PR as opposed to my assumptions.

The results here are quite different from the ones a few days ago. If I used the turnout of just 3 days ago for instance, I'd find that the 55+ were voting more and the Lower Mainland was lagging behind!

If I try to include more variables (percentage of people who have English as mother tongue or whether the seat was won by one or the other), I find the numbers of the right columns. It doesn't increase the R squared coefficient much (i.e: how much of the variation I can explain with these variables) and I find the same effects for the age or the regions. I do find that Liberals-held ridings are doing better than NDP-held ones. This isn't new and you have to wonder how much the NDP is getting its vote out.

Overall, the new regressions increase my confidence in my estimates above. The fact the previous turnout (in 2017) is a strong predictor means my age-turnout based on 2017 is most likely valid as base. After, we do observe the 18-34 and 55+ voting more than the 35-54. The net effect is unclear as the 18-34 are heavily behind PR whole the 55+ are in majority against it. There are more people in the 55+ group though. All in all however, the two might cancel out.

Remember that my vote estimates above are valid to the extend that the polls are right. In particular, the polls have to be right regarding the regional numbers as well as the levels of support by age. Additionally, these polls have found a large number of undecided and I'm assuming here that these undecided will either not vote or will break similarly to the decided voters. That'a big assumption and an error here (let's imagine undecided ultimately choosing the status quo) would make my estimates all wrong.


Conclusion

If the vote count was ending today, I think I'd bet on a small victory of the NO side. But with a full two weeks to go (note: Elections BC actually announced an extension to December 7th) because of the strike of Canada Post) and looking at the trend, I think proportional representation has a chance. It remains an incredibly close race and trying to predict the result is just a recipe to be ultimately wrong, but I'm okay with this! This is just occupational hazard for me at this point.

Can the 18-34 (in the Lower Mainland and on the Island) continue to vote so much and make the difference? Regionally, the large increase in the Lower Mainland is nice but I wish (as a PR supporter) I'd see ridings such as Point Grey or the two Victoria ones (on the island) getting higher turnouts than Quilchena for instance (Wilkinson's riding and the one that voted the most NO in the 2009 STV referendum).
The other day I was reporting that, based on the turnout data by riding provided daily by Elections BC, there were reasons to believe the NO side (i.e: keeping the current FPTP) was ahead. That was mostly because turnout data was indicating that older (55+) people were voting more, as well as higher turnouts in the interior. Those two groups are known (by the polls for instance) to be less favourable to PR.

However, the analysis was made difficult because by far the main determinant of the turnout was when the riding received the ballots. I was trying to account for this but nothing is ever perfect.

Today I decided to do a different exercise. I figured that turnout is nice (i.e: percentage of registered voters actually voting). But what ultimately matters is the number of votes. And we have that information!

We also know, thanks to the polls, the average level of support by age and by region. I can therefore use this information and try to estimate how many votes the YES and NO sides have received so far. This is obviously an exercise providing only estimates. The votes have not actually been counted. But I think it can be interesting.

The polls

The four polls (Angus-Reid, Insight West, Mainstreet and Research Co - I ignored my own Google Survey poll even though the results are completely consistent with the others) all mostly agree: the older you are, the less likely you are to support PR. Also, the Interior of BC is more pro FPTP than the Lower Mainland or the Island. They don't all agree on the actual numbers (it's normal) but overall, they show the same picture.

I thus averaged these four polls and used the averages to estimates how many votes have been casted for each side. Without going into too much detail, if a riding in the Interior has already mailed 4000 ballots and this riding has a higher proportion of 55+ than the rest of the province, my estimate will be that a majority of these 4000 votes are for keeping the current system. The exact number of votes for each side will be calculated based on the averages by age and region, as shown below.

Here are the averages I used.

PR FPTP
18-34 69% 31%
35-54 53% 47%
55+ 37% 63%


PR FPTP
Lower Mainland 52% 48%
Island 54% 46%
Rest of BC 43% 57%

Millennials and Gen Z are really, really in favour of PR while older Gen X and Baby-Boomers are really against it. Similarly, the rest of BC is really agaist the idea of switching system.

Estimates of the votes by day

Here are the total number of votes (they sum to the total provided by Elections BC, obviously) for each side. Yes it's close.



A better way to see who's ahead is to look at the percentages.



Here you can see that as the Lower Mainland ridings have started sending back their ballots (they got them later), the YES side has progressively climbed back. To the point where it could well be ahead right now.

Notice that this entire exercise is only valid to the extend that these polls are right. In particular, it is necessary to assume that the undecided (they are many of them in the polls) won't break in favour of one or the other side. Polls done over the last few weeks all showed the same 50-50 race so it doesn't look like the undecided are choosing one side more than the other so far.

Additionally, my estimates below are quite sensitive to the averages used. Adding the Research Co. numbers of yesterday made the NO side ahead at the beginning while it was behind the whole time before. At least no matter what averages I used, the trend has been the same: pro PR.

I tried to run these estimates using only the data on age or region. If I only use the age (and thus the age distribution profile in each riding), I would get the YES side at 50.78% as of today. Using only the region, I have the YES at 49.6% and therefore behind. It was at 48.7% at the beginning of the campaign.

So, why am I now saying the YES side might be ahead when my previous post was saying otherwise? I think the best explanation is that the regression was showing that indeed older people were voting more. That's still the case if I run the regression today. But the difference isn't big enough to compensate for the 18-44 that are more pro PR. Also, the influence of the Interior is vanishing as we keep counting votes. More importantly, it's one thing to show a higher turnout (so a higher %) but ultimately it's the total number of votes that will be counted. The Interior riding might be voting slightly more (note: as of today, my regression isn't showing any significant effect anymore regarding the region) but these ridings don't represent that many votes. Finally, and this is more stat nerdy stuff, but just because a result is statistically significant doesn't mean it's important in magnitude (it was a difference of less than 2% for the turnout in the Lower Mainland versus the Interior). In our situation, the difference in turnout by age (and therefore the bigger influence of the 55+) might not have that much of an impact.

Still, I think the conclusion remains that it's super close. The trend is good for the PR side and we still have many ridings that should provide many votes for a change that have a low turnout (the Victoria ridings for instance, some of the rare ridings that said YES in 2009). But similar to what the polls have been showing, this is a very close contest. The one thing that I still observe in my regressions is that ridings that voted more for the NDP in 2017 are not voting as much as others so far. If Horgan and the NDP are serious about wanting PR, they should be able to get the vote out in their own ridings.

I'll update this graph regularly (I don't promise to do it daily!).
This morning, as they do every day, Elections BC reported on how many ballots had been received and screened by them. The screening part is is important as Elections BC started reporting last week on how many ballots it has received in total. If the current turnout is only 10.6% if we look at the ballots processed, it's 21% of ballots received. A much higher number.

There is still considerable variations between ridings. The question is really: can these variations be interpreted in any way? For instance can we see a pattern more favourable to the YES or NO side? The main issue, as I previously reported, is that not all ridings received their voting packages at the same time. Some ridings, mostly in the interior, got theirs as early as October 24th while others, mostly in the Lower Mainland, had to wait until November 2nd. Even among the large group scheduled for the 2nd, you can still observe many differences.

The graph below shows you the turnout as a function of when the ballots were scheduled to arrived. As you can see, this is converging nicely.


Any analysis must account for this. The problem is that we don't have really detailed data. What we do have however is the turnout by day for each riding ever since Elections BC started reporting on them (the 5th). In this article, I use this information to account for when ridings received their ballots. Specifically, I look at when a riding crossed the 1% turnout and then count how many days it has been since. It isn't perfect but I think it's doing a pretty good job. I don't need a measure that is super accurate. What I need is a way to differentiate (or group) ridings in similar groups. So it doesn't matter if I estimate 3 days of voting and it was 6, as long as I'm off by the same margin for other ridings. I chose the 1% arbitrarily. I thought it was low enough that any riding that actually started voting should cross almost immediately.

Let me give you an example. The riding of Abbortsford-Mission only crossed the 1% mark on November 14th (I'm talking processed ballots here). There were ballots reported from this riding as early as the 5th (2 ballots specifically) but it took until the 14th to get enough ballots to represent 1% of registered voters. This riding was on the November 2nd schedule.

On the other hand the riding of Boundary-Similkamen crossed the 1% mark on November 5th, thus the first day reported by Elections BC. This riding was scheduled for October 25th.

You can therefore understand why, even today, the turnout in Boundary-Similkamen is 19.2% while it's only 6.3% in ABM. Of course the two should ultimately converge but it hasn't been the case yet.

And just to illustrate why my measure is better than simply using the scheduled date of Elections BC. Among the November 2nd ridings, 19 have been over 1% since the 14th while 12 ridings have crossed the mark on November 7th, 6 days before. Some ridings, such as Kelowna-Mission, were scheduled for November 2nd but actually crossed the 1% threshold on November 5th already. Thus, as you can see, there is considerable variation even among ridings technically in the same scheduled group. Differences can come from when they received the ballots or from a possibly selective processing from Elections BC (also, Canada Post could have its influence).

The graph below shows you the turnout as a function of how many days the riding had to vote (again, measured as how many days the riding has been over the 1% threshold).


A quick extrapolation would indicate that the final turnout should be around 30% if things continue at the same rate. Given that we are at 21% right now (received), 30% seems like a good guess unless it suddenly picks up in the last few days. I've been expecting a final result between 20 and 30% since the beginning. It seems I might have been slightly too conservative (or pessimistic).

This issue controlled for (well as much as I possibly can), we can try to look at other patterns. For instance is the Lower Mainland voting more or less than the interior? Are ridings with a higher proportion of 18-34 voting more? Of course we need to control for all other factors while doing this. This is why I used a regression. If you aren't familiar, think of it as a statistical tool looking at correlations but accounting for other factors at the same time. For instance, if I look at the simple correlation between turnout and age, some of the effects could instead come from the region. This is the case as ridings in the Lower Mainland have, in average, a younger electorate. A regression will be able to disentangle the two.

Without further ado, here are the results with the data of this morning:



There are two things you want to focus on. First of all is the sign of the coefficients. If it's positive (like with the share of 65+), it means the higher the share in a riding the higher the current turnout. Specifically, it means that if the share of people aged 65 and over is 10% higher, the turnout is 3.07% higher. For the number of days, you see that it's slightly less than 1% per day, which matches with the extrapolation of a final turnout around 30% so far given how many days there are left.

The second thing is the number (or lack of) of stars * next to the coefficient. If there is none, it means the relationship isn't statistically significant. To explain it in very simple terms, it means that we can't be very confident that there is indeed a relationship there. It could just be random noise from the limited sample we have. If there is no star, disregard the coefficient and imagine this is zero (i.e: no relationship between these variables). If there is one star, it's significant at 10% (so we are confident at 90%). 2 stars is 5% and 1 star is 1% or less (so certainty of 99% or more).

So, what can we learn? First of all, my measure of number of days above 1%, which is there to account for the fact some ridings have been able to return their ballots for a longer period, is by far the most important variable. Highly significant (t-stat of over 12 for my fellow nerds out there). What this means is variations across ridings is currently mostly caused by when the ridings received the ballots and/or how fast Elections BC has been at processing the ballots from this riding. In other words: I might be wasting my time here by using preliminary data.

We do observe a couple of other significant variables. So far it seems ridings with older people (over 55) vote more. Similarly the interior is voting more than the Lower Mainland (you need to interpret the coefficients for Lower Mainland and Island as differences with the Interior). Remember, in both cases this is while accounting for the fact many ridings in the interior got their ballots sooner.

All in all, this isn't looking very good for proportional representation. Polls have shown that younger people are more pro PR. Also, the interior is less in favour of PR than the Lower Mainland or Vancouver Island. At the same time, this is preliminary. Maybe younger, urban voters are taking longer to make their choice. But right now these effects are reinforcing each other and the odds of this referendum being successful aren't really great. The YES camp needs to step up its game and convince the younger voters (as well as the undecided) to mail their ballots back.


Edit:

Someone on Twitter suggested I instead looked at the change in turnout between last Friday and today as a way to determine where the new votes were coming from and possibly predict the future trend. I did run this regression. Without posting the full table, here are my findings:

Ridings that had fewer days of voting are catching up (the coefficient in front of number of days is negative). This is logical. Ridings with a higher proportion of 18-34 increased by more, so did the ridings with more 55+. It therefore seems it's really young people versus old ones and the middle age ones voting less (they most likely care less). Also, ridings with a higher share of votes for the NDP in 2017 have increased less than the others during this weekend. This is significant. Not sure why though but it seems to indicate the NDP isn't really getting its vote out.
The referendum on electoral reform is happening right now in BC. According to Elections BC, as of yesterday's morning, 6.5% of the registered voters have sent their ballots back (or, more exactly, Elections BC had received these ballots).

With every single poll showing pretty much a 50-50 race, including my own poll, turnout might be the key to this referendum. Which side will get its vote out the most? If Millennials and people on the Island vote the most, PR will pass. If, on the other hand, the votes come mostly from older people living in the interior, then the current system will prevail.

It's very early and therefore preliminary, but let's look at the numbers and see if we can find some patterns.

1. Turnout predictions

As mentioned above, the current turnout is 6.5% but there is considerable variation across ridings. This is mostly due to the fact that some regions (mostly in the interior of BC) got their envelops earlier. According to the website of Elections BC, there are four groups: 1 that received by October 24th, another by the 25, one by the 26 and, more importantly, a large group of 64 ridings that received their ballots until November 2nd. Among this group, you have variations as well (as indicated by the fact that some of these ridings were below 1% turnout even yesterday) but we don't have the exact date.

For us, it creates really two groups: the 23 ridings (16 in the interior, 4 in the Lower Mainland and 3 on the Island) that received early and the rest. It is important to keep that in mind during our analysis.

Elections BC started reporting on the number of ballots received on November 5th. For the first group of ridings, that means up to 12 days after receiving the envelops while the second group could have had as little as 3 days. We naturally expect the current turnout in the first group to be higher and it is. The average (not weighted, so each riding is equal) turnout is around 10.5% while it's only around 5% for the second group (another way to see the impact of the dates, the first group represents 36% of all ballots received while representing only 26% of the ridings).

This graph below shows you the turnout by date for the two groups:



As you can see the second group is catching up. As a matter of fact, at the same numbers of days since receiving the ballots, the second group is higher. But remember that the second group has more variety within it and not all ridings actually received their ballots as late as November 2nd. Also, comparing the number of days since received isn't truly an apple to apple comparison. According to Google Trends, the interest in this referendum picked up early November and peaked at the "lit" debate on November 7th (see graph below). This means that it's reasonable to assume the second group of ridings would have voted sooner while the first group wanted to wait. They should converge as we approach the deadline.



So, where is this heading? As some people right in fearing a record low turnout? Maybe.

Using this graph from David P. Ball from the Star, we see that the early turnout for this referendum is actually outpacing the previous two referendums (the 2011 one about the HST and the 2015 transit "plebiscite").


(Yes the period to return the ballots was longer for the HST referendum. This is because a postal strike happened in the middle of the campaign and Elections BC decided to extend the deadline. They might do the same this time around depending on what happens with Canada Post)

So, can we expect this referendum to break the 50% barrier? I'm skeptical.

I believe more people were interested in the HST (people actually requested this referendum after being upset at the BC Liberals for breaking their promise not to introduce this tax; Plus, you know, people really, really dislike sale taxes) or the transit plan than electoral reform. The Angus-Reid poll was clearly showing that for many people, this just wasn't a major topic.

Also, if I only use the data from the first group (the one that got their ballots earlier), we see that the best fit for the data seems to be a concave function (like a logarithm). The HST line on the graph above showed that turnout really picked up after around 16 days. If the same was to happen this year, the current data isn't showing this. But again, this is very early.

Let's do the extrapolation with multiple scenarios.

Scenario 1: we extrapolate the current concave function. In this case, we would predict a final turnout of around 20%.

Scenario 2: linear extrapolation. In this case, we end up at around 26%.

Scenario 3: it picks up. In this case, your guess is as good as mine. I can't really use the current data since it doesn't show an steepening of the slope. Based on the graph from the Star, 50% and more isn't impossible. Maybe a better way is to see where the HST turnout was at the same point as when this referendum will end. As you can see, this is just crossing the 30% mark.

Ultimately a turnout between 20 and 30% seems likely. Not only based on these graph but on the fact that during a similar referendum last year, Prince Edward Island got 36% turnout. If you account for the fact that PEI has higher general turnouts than BC, then 20-30% becomes quite good actually.


2. Any pattern?

Again, I repeat myself but this is very early and the data is not super reliable for now (the problem really is the second group of ridings that just started voting). Still, I decided to look at general correlations between the current turnout and some other variables such as the region, the electoral results in 2017 (is the riding a Liberal or NDP one?) as well as the median age and the share of Millennials in the riding (based on the 2011 census, haven't seen the updated version unfortunately).

What did I find? So far, it seems the Interior is voting more (and that's while factoring the fact most of the ridings in the first group were from this region; I used a regression to isolate each effect). For the age and shares of Millennials, they both have positive effects on the turnout! Is it possible that young people are voting (the 18-44, so that includes some Gen Z) and the older people also do? So a Baby-boomers versus Millennials while the Gen X would watch from the sideline. Finally, regarding who the riding voted for in 2017, I find that BC Liberals ridings are currently voting more (again, this is while accounting for when they received their ballots).

All of this is consistent with the Insight West poll showing that among the people who had voted already, FPTP was leading. Yes my regression is supposed to control for the fact some ridings received their ballots sooner, but it might not be doing a perfect job yet since we are too early in the process. Or turnout will simply remain higher among the BC Liberals, older ridings in the interior and proportional representation won't pass.

So right now it doesn't look too good for proportional representation. But we'll. My poll was clearly showing that the undecided were more similar to the pro-PR folks. The key now for Horgan and Weaver is to get these people (mostly Millennials) to actually vote.

As a conclusion, I think the key might be how well the Green party is mobilizing its voters. They are around 18% right now in the polls (so similar to last year). Their partisans are some of the most pro-PR, for obvious reasons. If the overall turnout does remain as low as 20%, then the Green vote could well be the difference. If this party can get its voters to vote, even half of them, that could be enough to reach a majority in favor of PR. The Liberals are definitely trying hard to get their vote out and I'm not sure about the NDP.
Just a few days after I published my own poll (conducted using Google Survey) on the referendum on electoral reform in BC, we now have a new Mainstreet poll. And it's showing exactly the same results!

Before talking about this poll, how remarkable is that? A $200 poll conducted using Google Survey can do just as good of a job as a full fledged, regular poll! You might not be excited about it but the political nerd/statistical geek in me is super excited. This is potentially revolutionary. Polling cost have already come down a lot in the last decade thanks to IVR and online sample, but Google Survey is just dirt cheap.

So, anyway, the new Mainstreet poll puts the "keeping the current system" at 50.5% among decided voter while "proportional representation" would receive 49.5%.

My own poll had 50.25% for keeping the current system (Note: yes when I published on Tuesday, "switching to PR" was slightly ahead -at 50.1%- but that's because I wanted to publish even though I still needed around 20 observations to be collected. Google Survey is great but it's crazy slow for the last few observations. It took almost a day and a half to collect these last 20 observations. When it was done, "keeping the current system" was back on top. It's irrelevant ultimately, with margins of error this was and is a 50-50 race).

The debate tonight between Horgan (NDP Premier of BC and pro-PR) and the Liberal leader Andrew Wilkinson (deeply against PR, as is most of the BC Liberals) will be interesting and incredibly important. Proportional representation was ahead in every poll done before this referendum. So Horgan and his partner, Green leader Andrew Weaver, are currently losing this campaign it seems.

Turnout so far has been... abysmal! Only 1.4% of registered voters have returned their mail-in ballot. Of course not everybody got their ballots at the same time and the postal strike doesn't help. Still, I wouldn't be surprised if the overall turnout, at the end, doesn't go above 30%.

It'll ultimately come down to turnout and undecided. Unfortunately Mainstreet doesn't indicate how many people are undecided. I'll ask them. [ Update: I asked and they didn't offer an undecided option at all; This goes a little bit against what I had seen before where the question without undecided would be more pro-PR than the question with the option]. The poll also confirms what my own poll was showing: people under 35 are the biggest supporters of proportional representation.

The Mainstreet poll also confirms what is common sense: if we switch to PR, Multi-Member proportional is the preferred system. Maybe surprisingly, the never used before Dual member proportional is third while the weird urban-rural proportional is second. I would have expected the opposite.
Full disclosure: I voted to switch to proportional representation. I have never hidden the fact I was pro-PR even if adopting PR pretty much puts me out of a job since projections wouldn't be as important anymore. Just because I have an opinion on this topic doesn't mean I can't be impartial when doing an analysis.

According to a poll conducted exclusively for this blog by Google Survey, British Colombians are almost perfectly divided on the electoral reform referendum. An important share of the population is however undecided and could ultimately make the difference.

Even though we are in the middle of the referendum period (many people, including myself, have already returned their mail-in ballot) and the debate between the NDP-Premier John Horgan and  Liberal leader Andrew Wilkinson is this Thursday, we've had almost no poll. The only one is from over a month ago! So I decided to spend some of that Google ads money my site generated during the Quebec election and to order a poll from Google Survey (I guess Google is winning at the end). I had already tried for the Ontario election with good results (especially in hindsight since most polls underestimated the PC and overestimated the NDP, although my poll was done a little bit far from the election date, so I might have been lucky). I also ordered a "fun" poll asking people what they thought of Maxime Bernier's infamous tweet. Google Survey has also been used in the US with great success for electoral polling.

Google Survey uses the usual Google database (usually used for marketing purposes) to create a representative sample of the Canadian population. They are fairly cheap (20 cents per answer as long as you only ask one question; It increases to $1 and more if you want multiple questions). This poll was done from October 31st to November 6th (so right after the peak in interest for this referendum, thus when people would likely start voting; They have until the end of the month to do so). The sample size is 1200 respondents, although the final number of observations is just over 800 because many respondents didn't have an age and/or gender and were therefore dropped (given a weight of zero). The specific question I asked was "There is currently a referendum on electoral reform in BC. You have or will receive your voting package in the mail. Which following option do you intend to vote for?". I only asked this one question and didn't try to replicate the format of the referendum where a second question exists (you have to choose between 3 potential proportional systems). I did it for monetary reasons and because, quite frankly, the main question is whether BC will switch or not. No actual margins of error since the poll is online but a probabilistic sample of that size would have margins of 3.4% 19 times out of 20.

Here are the results.


First observation: it's close! So, so close. Among decided voters, it's essentially a 50-50 race. And it has been like that all along the period during which the data was collected. "Keeping the current system" was actually ahead for the most part until the last 50 observations. This referendum doesn't have a qualified majority threshold, so 50%+1 would be enough.

Notice however the very high number of people who are undecided. I wish I had the budget to ask them a leaning question but I didn't. I'm not surprised by the high number though. Electoral reform can be seen as a complex issue, or one that doesn't matter. It also seems that online polls, and Google Surveys in particular, have a higher share of undecided. It's interesting to see more people being undecided on this topic than regarding Bernier's tweet!

Let's compare these numbers to the Angus-Reid poll of September. At that time, 33% were undecided, 31% in favour of keeping the current system and 33% wanted to switch. This meant that among decided voters, it was 52-48. My poll numbers match pretty well (not withstanding the higher number of undecided). Can we take it as evidence of a negative trend against proportional representation? Not really since once you apply the margins of error, both polls pretty much match.

With that said, we had seen a trend over the last year according to Angus-Reid. I mentioned it in my post about that poll. It's not clear if Horgan and Weaver are currently losing their referendum or if it's more a question of allowing a "not sure/Don't know" option in the poll. It is possible that pro-PR people are simply less sure of their choices than anti-PR ones. So the 40%+ of undecided in this poll might actually break slightly in favour of switching (note: I expect most of them not to vote at all). The Tyee wrote about comparing different polls with different methodologies already.

Still, I think it's fair to say that this referendum could be very close and I'm not sure the YES campaign is winning so far. Fear mongering arguments (proportional representation will literally allow a nazi party to have seats!...) and others, more valid arguments (the systems offered are half defined, in particular MMP where they haven't even chosen what kind of lists we'd use!) can easily convince some voters against taking the risk to switch. If Horgan is serious in his desire to adopt proportional representation, he'll need to give a convincing performance on Thursday. I'd suggest to maybe emphasize more on the fact that we'd vote again after 2 elections to confirm we want to keep PR. Also, I maintain my position that this referendum would be an easier sell if we could all agree that only MMP is a viable option (Dual member proportional is a system never applied anywhere, invented by a student in mathematics in Alberta and is, ultimately, just a modified version of MMP with more complex features such as vote transfers and forfeited seats; Urban-Rural proportional is too complicated -it mixes STV and MMP- and I can't see it winning). Doing so could clarify the choice for people. Just my 5 cents of course.

Ok, let's look at more numbers since, after all, I paid for this poll and I'm going to milk it. First, by age:



It seems to be Millennials vs Baby boomers (and older Gen X). The general trend isn't surprising but the extend of the difference between age groups might be bigger than expected. Millennials and later Gen X really, really support proportional representation while older Gen X and Baby Boomers, really don't.

The 18-24 is a little bit surprising and might be due to an overall small sample size (or way too many undecided, see below).

Finally, if I look at the age profile of the undecided, I find that they are younger than the population in general. For instance, over 25% of the undecided are between 18-24 while they only represent 20% of the total sample. On the other hand, only 5% of the undecided are aged 65 and over (versus over 8% of the sample coming from this age group).

Here is the percentage of undecided by age.


The fact the undecided are mostly in the age groups that are otherwise favourable to proportional representation is potentially good news for John Horgan and Andrew Weaver. But it also means they need to make sure to convince them to vote ultimately. Turnout will likely be very low (Prince Edward Island got 36% during a similar mail-in referendum last year but this province is used to turnouts much higher than in BC). So every vote could be very important. Now let's wait and see if the debate this Thursday will change this referendum.

It'd also be interesting to reweigh the data based on the age distribution of BC voters as opposed to the census. But that will be for another day for me.

If you'd like to have access to the raw polling data, email me your request along with the reason 9and who you are!).