FAQ
What is a Rep Order? What does TO (or Tr) stand for?
Please see the Pundit Metrics page for a glossary and full definition of all the electoral terms and measurements used in the Pundits' Guide.
Est ce-que cet site web est disposible en français?
Le base de données est déjà traduit, mais je n'ai pas encore traduit les pages web. J'espère accomplir ce travail bientôt.
Why are there no results before 1997?
Because the data for the earlier elections is not readily available in digital form (or, if it is, I haven't found it yet). I would like to add it in, not least because this was the period in which I was more actively involved in election campaigns myself, and am nearly done with the manual data entry.
Do you have poll-by-poll results?
Here was my old answer: "No, because the poll boundaries usually change from election to election and you need a lot of local knowledge, or a GIS database not currently in the public domain, in order to make the transpositions properly … not to mention a LOT of local knowledge in order to interpret them properly. That said, I don't know an organizer who wouldn't salivate at the prospect of having this data available at the click of a button. Sigh."
Several years later, I've managed to assemble them all into a relational database from which I draw the "Poll-by-Poll Metrics" on a riding page. The polling division boundaries are more stable than I thought, but are still not guaranteed to be identical over time, but I'm working on understanding this data better, and having access to poll-by-poll Google Maps is going to help with that a lot.
Why didn't you show the results for *all* the political parties?
Fair question. And, in fairness, for some of the result tables it would have been prohibitive to lay them out to show all the parties, so when the project was just for personal interest I grouped all other parties under "Other" in the database itself, rather than in the query. Over time I will try to add this data, especially since there are some candidates who have run for both the smaller and larger parties.
How about provincial results?
Well, you certainly have identified all the items on my long-term to-do list now, haven't you.
Hey, I bookmarked a page, but when I went back it was different from what I bookmarked. What gives?
This site uses Ajax to dynamically change its content. You probably did some additional searches after getting to that page. Click on the "Permalink" hyperlink before bookmarking any search and you should be fine. If that still doesn't work, please contact me through the Contact page, because either you're using a really old browser that doesn't support Ajax, or else I probably screwed up somehow.
[NOTE: During the transition period from Blogger to WordPress, the "Permalink" feature is temporarily unavailable.]
Sheesh, there's a glaring error in your data. Don't you know anything? <or> This site doesn't work properly/the way I expected.
Well, then please let me know right away, either using that page's "Report an error" hyperlink or through the Contact page, because lord knows I am fallible (with a capital F) and could use the help of an army of proof-readers.
We recently had a lecture at work about how the power of social networking on the Internet has allowed archives of knowledge to be self-correcting within about 20 minutes. That's kind of cool if you think about it. So, now, do your bit and report the error.
Man, don't you even know about news-event-x that happened this morning? How often do you update this thing anyways?
At the moment, I'm trying to trade off new development against updating the data, and balance off these two priorities against each other … AND THE REST OF MY LIFE. I'm a public servant, remember? If I updated this thing all day, you'd probably be the first one reporting me for uploading during work hours. Why not just drop me a line instead, and let me know?
How in the heck do you read those charts? They look like 'Voice of Fire'.
Ah, you must be talking about the one with a stacked bar chart sitting on top of a stacked area chart. I needed to find something that was the equivalent of a pie-chart that changed over time, and that was one of the only two available choices. Think about the plot-area of the area chart as the total electorate, and see how each party's share of the vote changes over time (and which other party it's coming from or going to) … then compare it to how many seats the party won with that percentage of the vote (shown in the bar chart).
PCs and Conservatives are both blue, by the way; the Canadian Alliance and Reform parties are grouped together and have an aqua colour. The Green party has a more yellowy green colour, and usually only appears later on in any strength. Honestly, I couldn't read them at first myself, but now that I've got the hang of it, I can actually glean quite a bit of info about a region just by looking at the chart.
Anyways, after enough people complained about them, I found the other solution which has become the default chart display. The 'Voice-of-Fire' charts have become the 'Expert Charts' … to see if I can at least generate a bit of snob appeal that way.
Why didn't you add together the results of PartyX and PartyY to show the impact of strategic voting?
I guess I'm just trying to let the numbers speak for themselves. Whether strategic voting has an empirical basis, or is simply an electoral strategy of one or other of the political parties; whether it has ever met its stated (or unstated) objectives, or whether actual voters will behave the way pundits think they ought to — well I leave that for you to decide. My goal here is simply to present the kind of data you can use to make that assessment for yourself.
However, if you want a starting point for this kind of analysis, browse results by Election, and then use the Pundit Query that shows ridings won with less than 50% of the vote, paying attention to the %Margin. Or, still browsing by Election, use the Pundit Query that shows ridings which changed hands (party-wise) during that campaign. Don't forget to factor in any incumbency issues and changes to the turnout and number of electors to your analysis, though.
Where did you get the regional groupings of ridings?
Made them up myself, basically. The objective was to group ridings according to commonalities in voting patterns, demographics, geography, community and media markets … factors that could all arguably affect the results … so that's why sometimes there are regions made up of only one riding (for e.g., Labrador), or why I split GTA-Toronto into North and South regions, rather than East and West as used by other sites. I am woefully ignorant of any of this detail in the provinces of Québec and New Brunswick, so would really appreciate any suggestions for improving the groupings there.
How come you don't say where you were involved in election campaigns, or which party you worked for?
Anyone who can find this site can probably find out that for him- or herself if they try hard enough. The thing is, anyone who has worked on election campaigns *also* gets the part about how the party that wins the most seats at the end, and enjoys the confidence of the House of Commons, gets to run things and deserves the loyalty and best advice of a professional public service. Which is why I'm not telling, at least not here.
I've tried hard to keep this site scrupulously non-partisan and balanced. I hope the numbers allow pundits, both amateur and professional, from all parts of the political spectrum to assess the parties' spin, the journalists' reporting, and the bloggers' … um … unbiased analyses.
Happy punditry, everyone !
