4 ms·

Designer fixed U.S. election maps

8 points by colinprince 6y ago

srtjstjsj 6y ago
Mark Newman's team did this at least 8 years ago with cartograms, but the general public doesn't care. Lots of free code is available. http://www-personal.umich.edu/~mejn/election/2016/ http://www-personal.umich.edu/~mejn/election/2016/ https://newrepublic.com/article/115550/cartograms-are-important-inforgraphic-tool https://newrepublic.com/article/115550/cartograms-are-import...
- bobthepanda 6y ago
  That specific style of cartogram is pretty bad because it distorts shapes in quite odd ways and humans aren‘t that good at judging spatial relativity. I prefer hex maps. http://blog.apps.npr.org/2015/05/11/hex-tile-maps.html http://blog.apps.npr.org/2015/05/11/hex-tile-maps.html
  - lambertsimnel 6y ago
    It would be interesting to see a hex map that represented each state with a number of tiles in proportion to its electoral college votes (or in some other way showed their relative weight). There's one for last year's UK election here (but it doesn't reflect margins of victory): https://www.theguardian.com/politics/ng-interactive/2019/dec/12/uk-general-election-2019-full-results-live-labour-conservatives-tories https://www.theguardian.com/politics/ng-interactive/2019/dec...
    - bobthepanda 6y ago
      NYT has a map showing the states as squares with roughly the same concept. Generally speaking, hex maps work a lot less well for the US because you have to pick between either extremely heavy distortion (to the point of meaningless abstraction), or a massive gap in the Plains and Rockies. https://simonrogers.net/2016/09/22/tilegrams-make-your-own-cartogram-hexmaps-with-our-new-tool/ https://simonrogers.net/2016/09/22/tilegrams-make-your-own-c... The UK's population density is not quite so comically varied; a hex map of the UK still kind of resembles the UK.
      - lambertsimnel 6y ago
        Thanks for your reply. The link especially was fascinating.
lm28469 6y ago
> I had to correct this visual mistake It's not a mistake, the first map represents something and the second represents something entirely different.
- eyelidlessness 6y ago
  But the binary maps are easily misleading (and can easily be used to intentionally mislead). Perhaps they shouldn’t be the default for communicating electoral success.
  - danielscrubs 6y ago
    I’m not an American so correct me if I’m wrong but winning the popular vote doesn’t mean success either in US if I understand it correctly, so this wouldn’t solve the problem at hand right? U.S. Constitution doesn't require electors to follow their state's popular vote so the map would also have to reflect that for certain states.
    - bobthepanda 6y ago
      You could still apply the same principle to electors. http://blog.apps.npr.org/2015/05/11/hex-tile-maps.html http://blog.apps.npr.org/2015/05/11/hex-tile-maps.html NPR‘s front page election map also has a map with scaled squares representing states.
      - danielscrubs 6y ago
        That seems better to me than original posters map
        eyelidlessness 6y ago
        Each provides completely different data meant to communicate different things. The point of the (first) original map communicates winning party geographically by (I believe) county. The “fixed” map communicates relative share of vote by population, because the designer rightly objects to the use of that geographical data to vastly overstate one party’s political support. The map here communicates electoral outcome. None of them is objectively “better” although one is especially misleading when misused. The “better” you’re seeking/identifying is better for an entirely different analysis than what was intended. The “fixed” version doesn’t necessarily correct any misunderstanding of the electoral system, but it certainly better presents public political distribution than either of the others. If the intent is to capture both public support and electoral outcome you actually need more axes, and probably the ability to toggle different presentations of the same data. A “better” visualization capturing the designer’s intent and yours as I understand it would present: 1. Distribution of votes by count, not by arbitrary boundaries (to capture public sentiment). 2. Weight of vote by geography (to capture electoral outcome). 3. Some way to quantify the discrepancy between 1 & 2, and ideally some additional text resources discussing the historical background behind that discrepancy and its impact on democracy.
arethuza 6y ago
Similar kinds of maps have been used in the UK as well - obviously not as extreme a case as the US but still interesting: https://geographical.co.uk/geopolitics/geopolitics/item/3557-ukge2019 https://geographical.co.uk/geopolitics/geopolitics/item/3557...

8 points by colinprince 6y ago

Mark Newman's team did this at least 8 years ago with cartograms, but the general public doesn't care. Lots of free code is available. http://www-personal.umich.edu/~mejn/election/2016/ http://www-personal.umich.edu/~mejn/election/2016/ https://newrepublic.com/article/115550/cartograms-are-important-inforgraphic-tool https://newrepublic.com/article/115550/cartograms-are-import...

bobthepanda 6y ago
That specific style of cartogram is pretty bad because it distorts shapes in quite odd ways and humans aren‘t that good at judging spatial relativity. I prefer hex maps. http://blog.apps.npr.org/2015/05/11/hex-tile-maps.html http://blog.apps.npr.org/2015/05/11/hex-tile-maps.html
- lambertsimnel 6y ago
  It would be interesting to see a hex map that represented each state with a number of tiles in proportion to its electoral college votes (or in some other way showed their relative weight). There's one for last year's UK election here (but it doesn't reflect margins of victory): https://www.theguardian.com/politics/ng-interactive/2019/dec/12/uk-general-election-2019-full-results-live-labour-conservatives-tories https://www.theguardian.com/politics/ng-interactive/2019/dec...
  - bobthepanda 6y ago
    NYT has a map showing the states as squares with roughly the same concept. Generally speaking, hex maps work a lot less well for the US because you have to pick between either extremely heavy distortion (to the point of meaningless abstraction), or a massive gap in the Plains and Rockies. https://simonrogers.net/2016/09/22/tilegrams-make-your-own-cartogram-hexmaps-with-our-new-tool/ https://simonrogers.net/2016/09/22/tilegrams-make-your-own-c... The UK's population density is not quite so comically varied; a hex map of the UK still kind of resembles the UK.
    - lambertsimnel 6y ago
      Thanks for your reply. The link especially was fascinating.

lm28469 6y ago

> I had to correct this visual mistake It's not a mistake, the first map represents something and the second represents something entirely different.

eyelidlessness 6y ago
But the binary maps are easily misleading (and can easily be used to intentionally mislead). Perhaps they shouldn’t be the default for communicating electoral success.
- danielscrubs 6y ago
  I’m not an American so correct me if I’m wrong but winning the popular vote doesn’t mean success either in US if I understand it correctly, so this wouldn’t solve the problem at hand right? U.S. Constitution doesn't require electors to follow their state's popular vote so the map would also have to reflect that for certain states.
  - bobthepanda 6y ago
    You could still apply the same principle to electors. http://blog.apps.npr.org/2015/05/11/hex-tile-maps.html http://blog.apps.npr.org/2015/05/11/hex-tile-maps.html NPR‘s front page election map also has a map with scaled squares representing states.
    - danielscrubs 6y ago
      That seems better to me than original posters map
      - eyelidlessness 6y ago
        Each provides completely different data meant to communicate different things. The point of the (first) original map communicates winning party geographically by (I believe) county. The “fixed” map communicates relative share of vote by population, because the designer rightly objects to the use of that geographical data to vastly overstate one party’s political support. The map here communicates electoral outcome. None of them is objectively “better” although one is especially misleading when misused. The “better” you’re seeking/identifying is better for an entirely different analysis than what was intended. The “fixed” version doesn’t necessarily correct any misunderstanding of the electoral system, but it certainly better presents public political distribution than either of the others. If the intent is to capture both public support and electoral outcome you actually need more axes, and probably the ability to toggle different presentations of the same data. A “better” visualization capturing the designer’s intent and yours as I understand it would present: 1. Distribution of votes by count, not by arbitrary boundaries (to capture public sentiment). 2. Weight of vote by geography (to capture electoral outcome). 3. Some way to quantify the discrepancy between 1 & 2, and ideally some additional text resources discussing the historical background behind that discrepancy and its impact on democracy.

arethuza 6y ago

Similar kinds of maps have been used in the UK as well - obviously not as extreme a case as the US but still interesting: https://geographical.co.uk/geopolitics/geopolitics/item/3557-ukge2019 https://geographical.co.uk/geopolitics/geopolitics/item/3557...