Wikipedia:WikiProject Electoral districts in Canada/Demographics

From Wikipedia, the free encyclopedia

This section is intended to present census data and whatever miscellaneous info on the character that can be presented as a number. Three big questions need to be answered: What data? Data from what year? How to present the data?

Data options: population, electors, population pyramid, income, population density, area of district, population change, average age, education level, homeownership rate, vacancy rate, religion, ethnic background, ...

Year options: last census year (2001), last election year (2004), trend over district's history,....

  • Population change worked really well in the B.C. provincial electoral districts but may not be available for the federal districts. Also, how does your choice of census data differ from one district to another, and what does that mean? Is "Demographics" really the best title for this section?


  • In my opinion the current system for displaying demographic information isn't very good. There are already too many tables, so adding yet another is a bit much. The small left aligned box also generates an excessive amount of whitespace. I would prefer if this information were merged into the general riding infobox. - SimonP 22:49, 25 October 2005 (UTC)
  • Let's choose the information we want to present, then figure out how to present it. Personally, I think some of this information belongs in the infobox for election results, since the data changes with each election. The demographics section should contain info for the most recent election (or an upcoming one?), and have a much broader scope. Mindmatrix 21:08, 27 October 2005 (UTC)


Contents

[edit] Which data?

Make a proposal and defend its relevancy. Vote.

  • Population. Any objections?
    • Gross population pyramid (eg: 0-17, 18-35, 36-54, 55+); this also gives us voting-age residents (but not electors). Large deviations from Canadian/provincial averages should be noted. Mindmatrix 15:40, 1 November 2005 (UTC)
  • Population Change, 1996-2001. Reveals population and economic trend. Voting trends may emerge from areas that are growing and others that are contracting. --maclean25 06:12, 28 October 2005 (UTC)
    • This could be placed in the same table as population, if we have a data source for 1996 census data in 2003 FED Representational Order. I have yet to find such a table on the StatCan website, though. Mindmatrix 15:40, 1 November 2005 (UTC)
  • Electors. Useful for checking voter-turnout calculation. --maclean25 06:05, 28 October 2005 (UTC)
    • Is there a good online source for this in (X)HTML or XML format? (A queryable database is good too.) Mindmatrix 23:10, 3 November 2005 (UTC)
  • Income and Work. Include only participation rate, employment rate, unemployment rate and average income (individual, household and/or family). Compare to province and nation. Mindmatrix 15:40, 1 November 2005 (UTC)
    • Support unemployment rate as a graph. Oppose average income as being easily distorted. Neutral on others because I don't want to bog down this Demographics section. (added ext links to Mindmatrix comment above) --maclean25 08:41, 3 November 2005 (UTC)
      • When you say as a graph, do you mean tracking monthly/yearly changes? If so, someone would have to keep them up-to-date. I'd rather not bog down the Demographics section either, but I think Income and Work is one of the most-cited stats about a riding, after population, which is why I proposed its inclusion. Mindmatrix 23:10, 3 November 2005 (UTC)
  • Families and dwellings. Include only family structure (married, common-law, lone parent), average dwelling value, and ownership/rental ratio. Compare to provincial and national averages. Mindmatrix 15:40, 1 November 2005 (UTC)
    • Oppose family structure (relevency to voting?) --maclean25 08:41, 3 November 2005 (UTC)
      • For Toronto, there are some areas with a high lone parent ratio, almost all of whom rent rather than own. This tends to affect voting behaviour in the riding. These areas tend to be left-leaning politically. Mindmatrix 23:10, 3 November 2005 (UTC)

[edit] Suggested

  • Religion. Should this be for those ridings which deviate from the typical situation (close split between Catholic and Protestant, with significant non-religious groups), or just do it for all of them? Which data do we present: all religious groups, the 3/4/5 most common denominations in the riding, or something else? Mindmatrix 15:40, 1 November 2005 (UTC)
    • Question does this affect voting patterns (like Conservative vs. Liberal?, or even Christian Heritage?). --maclean25
      • There are some observable trends - strongly Protestant/evangelical areas tend to vote Conservative, Roman Catholics tend to vote Liberal in Ontario and are left-leaning in Quebec, and some ridings are affected by religious split. I'm not sure how strong the trends are though. Its difficult to assert the influence of religion, or any other factor, on riding results, since there are relatively few elections and far too many external factors. (Statistically, its difficult to determine the correlation.) Mindmatrix 23:10, 3 November 2005 (UTC)
        • But I don't think the available data permits making that distinction between evangelicals and mainstream Protestants. They are all ither under "Protestant" or "Christian". The Catholic numbers are essentially meaningless where 83% of Quebeckers report themselves as Catholics, but very few let that affect their attitudes towards a number of social issues (marriage, abortion, etc.), orthe way they vote. Luigizanasi 08:11, 7 November 2005 (UTC)

[edit] Other considerations

Note: These will not likely be implemented unless strong arguments can be presented for their inclusion.

  • Aboriginal identity. Perhaps only in ridings where this is relevant? (how do we define relevant for this situation?)
    • Oppose, inconsistent (and flawed) question over censusesses (censi?), unless the Aboriginal population in New Brunswick actually increased 66% between 1996-2001. --maclean25 08:41, 3 November 2005 (UTC)
      • Support. While there are problems with aboriginal self-identification, changing attitudes have changed people's responses to this. The same was true of German origins which mysteriously declined after the 1st & second World Wars. Or look at the number of people who report "Canadian" as their ethnic origin (even in Quebec) despite being explicitly asked where their ancestors came from. So the ethnic origin questions, relying on self reporting are all ineherently flawed, but the best we have. In any case, it is what people are comfortable reporting themselves as, so they actually reflects the social reality. Luigizanasi 08:07, 7 November 2005 (UTC)
  • Visible minorities. Even better would be a breakdown by ethnic group, but StatCan does not provide this in the standard tables.
    • "Visible minorities" and "immigrant population" sort of give the info. Luigizanasi 08:07, 7 November 2005 (UTC)
  • Education level.

How about percentage with university education? Luigizanasi 08:07, 7 November 2005 (UTC)

  • Mother tongue. Unfortunately, StatCan only list English, French, other or multiple.
    • Support for Quebec only. Language does not manifest itself in voting pattern in any other province. --maclean25
      • Woodbridge is overwhelmingly Italian, and has consistently voted Liberal. Of course, this is probably not due to language, but rather ethnicity or religion. Since StatCan doesn't list non-official languages anyway, my point is moot. Does language factor into New Brunswick election results? Mindmatrix 23:10, 3 November 2005 (UTC)
      • I thought that the high francophone population ridings in Northern Ontario tended to vote Liberal or NDP, so Support. Luigizanasi 08:07, 7 November 2005 (UTC)

[edit] Data from years?

Last census year? Election year? Range of years (eg. in a graph)?

We're restricted to the data that's available to us, so most of it will be from the last census, and elector counts etc. will be from the most recent election. I think we should restrict the Demographics section to only the most recent information, since it changes considerably over time. Historical demographic information of relevance can (should?) be added to the appropriate election results section. We could, of course, have a demographic trends section too... Mindmatrix 23:16, 3 November 2005 (UTC)

[edit] Format

Table? List? Infobox? Graph? Illustration? Pie Chart? Please provide an example in a sample electoral district of your choice.

Here is a proposed table. Needs to be prettified, obviously. Note the right alignment. Also, the table entries should be right aligned. These are the figures I consider probably significant. The popualtion by age could be percentages. It is hard to get significant numbers for religion, as the evangelicals, etc. are mosly under "Protestant" and "Christian". Feel free to fix. Substantial changes should probably be presented as another table, so we can choose. Luigizanasi 07:51, 7 November 2005 (UTC)

Riding profile
  Yukon
electoral

district

 Canada
average
Population (2001) 28,675 97,426
Population 18+ 23,510 81,475
Population 60 years + 2,700 16,761
Per cent English mother tongue 86.2% 58.5%
Per cent French mother tongue 3.1% 22.6%
Per cent other mother tongue 9.5% 17.6%
Percent immigrants 10.5% 18%
Per cent visible minorities 3.6% 13.4%
Aboriginal population 22.9% 3.3%
Per cent with university education 18.6% 17.9%
Average household income $60,236 $58,360
Unemployment rate 11.6% 7.4%
Per cent home owners 62.9% 65.8%
Average dwelling value $140,447 $162,709



  • I like the idea of having the demographics off to the side. Perhaps all the demographics that consist of a single variable (ie. not a graph) can be placed in an infobox. However, we run the risk of having the box run longer than the article. Please see Langley (electoral district) for my selection of demographics. I really like the idea of comparing it to a national or provincial average. --maclean25 08:31, 10 November 2005 (UTC)