10-31-2019, 06:13 PM
Cross-posted from the old forum:
https://www.tapatalk.com/groups/the_coff...ml#p346931
I'm a big fan of graphs, charts and tables: they're often great at presenting data in a clear and concise format, and besides, graphs and charts are often visually interesting (tables less so, but they're still useful ). Unfortunately, these visual aids also end up being abused, in such ways that they just mislead people . Sometimes, it's done deliberately (for example, with a crafty choice of the scale); other times, it's done accidentally (due to a poor understanding of the underlying mathematics).
I'll start off with an example that was put together by the Liberal Democrats ahead of the forthcoming UK general election. This graph implies that their candidate is only 6 points behind their Conservative opponent - and therefore, if the other parties got on board with the Lib Dems, then they could flip the seat:
Trouble is, the poll didn't ask "Who do you intend to vote for?", as is being implied. Instead, if you read the fine print, you'll see that it was asking "Imagine that the result in your constituency was expected to be very close between the Conservative and Liberal Democrat candidate, and none of the other parties were competitive. In this scenario, which party would you vote for?". That's a very different question - and I'd have to imagine that the 8% who would still vote Labour under that scenario are absolute die-hards, and not easy targets to be flipped by the Lib Dems!
So, do you have any other bad or misleading graphs, charts, or tables to share?
https://www.tapatalk.com/groups/the_coff...ml#p346931
I'm a big fan of graphs, charts and tables: they're often great at presenting data in a clear and concise format, and besides, graphs and charts are often visually interesting (tables less so, but they're still useful ). Unfortunately, these visual aids also end up being abused, in such ways that they just mislead people . Sometimes, it's done deliberately (for example, with a crafty choice of the scale); other times, it's done accidentally (due to a poor understanding of the underlying mathematics).
I'll start off with an example that was put together by the Liberal Democrats ahead of the forthcoming UK general election. This graph implies that their candidate is only 6 points behind their Conservative opponent - and therefore, if the other parties got on board with the Lib Dems, then they could flip the seat:
Trouble is, the poll didn't ask "Who do you intend to vote for?", as is being implied. Instead, if you read the fine print, you'll see that it was asking "Imagine that the result in your constituency was expected to be very close between the Conservative and Liberal Democrat candidate, and none of the other parties were competitive. In this scenario, which party would you vote for?". That's a very different question - and I'd have to imagine that the 8% who would still vote Labour under that scenario are absolute die-hards, and not easy targets to be flipped by the Lib Dems!
So, do you have any other bad or misleading graphs, charts, or tables to share?
Board Information and Policies
Affiliation | Coffee Credits | Ranks and Awards | Name Changes
Account Deletion | BBCode Reference
Moonface (in 'Woman runs 49 red lights in ex's car')' Wrote: If only she had ran another 20 lights.
(Thanks to Nilla for the avatar, and Detective Osprey for the sig!)
My Items