3 ms·

they pretend to convey unbiased information

by joubert 17y ago

hugh3 17y ago
Indeed. The infographic about which news source's audiences are better informed is a pretty good example of "how to distort with infographics": http://www.wired.com/culture/culturereviews/magazine/15-07/st_infoporn http://www.wired.com/culture/culturereviews/magazine/15-07/s... Three big problems: 1. On the left, did they really only ask three questions in an attempt to find out who was better informed? If not, what happened to the other questions that (perhaps) didn't support their point? If so, why did they only ask three questions when they could have asked more? Also, the wording isn't specified; what does it mean to be able to "identify Scooter Libby"? 2. The area of those bubbles is not proportional to the number inside. I'm pretty sure they've made the radius proportional to the number, so the area (and psychological bigness) is proportional to the square of the number, thus exaggerating relatively small differences. 3. On the right hand side, the questions are arranged so as to decrease the visual significance of the four questions where knowledge increased from 1989 to 2007 and increase the visual significance of the four questions where it has decreased. Then, despite the fact that knowledge has gone up in four cases and down in four, they stick the big headline "Americans know less about politics than they did in 1989" at the top and you have to look very carefully at the data to realise that it's pretty much a wash. Anyway, while infographics can be useful, you want to be very careful in reading them. Ask yourself: "Why would someone go to so much trouble to present this information visually if they weren't trying to distort it in some way?"
jaybol 17y ago
I understand that images and visual representations are powerful, and impact more people in large part because they require less in terms of time/attention, so they have the power to mislead. But to label all infographics as a scourge and as automatically biased is quite a generalization. For that matter, how do they claim to be unbiased in the first place? Data and studies are spun towards researcher bias at times regardless of the end format. To disregard infographics or despise them versus black and white text and Excel tables doesn't make sense. It is a medium, and a powerful medium at that, but so is text, photography and videos, and they can all be used to objective ends or subjective ends, and none is ever pure and holy or automatically accurate (nor are Excel tables and the raw data used to inform the stylized infographics).
- joubert 17y ago
  Text may be black and white physically, but can express a wider range of nuance.

they pretend to convey unbiased information

by joubert 17y ago

they pretend to convey unbiased information

hugh3 17y ago

Indeed. The infographic about which news source's audiences are better informed is a pretty good example of "how to distort with infographics": http://www.wired.com/culture/culturereviews/magazine/15-07/st_infoporn http://www.wired.com/culture/culturereviews/magazine/15-07/s... Three big problems: 1. On the left, did they really only ask three questions in an attempt to find out who was better informed? If not, what happened to the other questions that (perhaps) didn't support their point? If so, why did they only ask three questions when they could have asked more? Also, the wording isn't specified; what does it mean to be able to "identify Scooter Libby"? 2. The area of those bubbles is not proportional to the number inside. I'm pretty sure they've made the radius proportional to the number, so the area (and psychological bigness) is proportional to the square of the number, thus exaggerating relatively small differences. 3. On the right hand side, the questions are arranged so as to decrease the visual significance of the four questions where knowledge increased from 1989 to 2007 and increase the visual significance of the four questions where it has decreased. Then, despite the fact that knowledge has gone up in four cases and down in four, they stick the big headline "Americans know less about politics than they did in 1989" at the top and you have to look very carefully at the data to realise that it's pretty much a wash. Anyway, while infographics can be useful, you want to be very careful in reading them. Ask yourself: "Why would someone go to so much trouble to present this information visually if they weren't trying to distort it in some way?"

jaybol 17y ago

I understand that images and visual representations are powerful, and impact more people in large part because they require less in terms of time/attention, so they have the power to mislead. But to label all infographics as a scourge and as automatically biased is quite a generalization. For that matter, how do they claim to be unbiased in the first place? Data and studies are spun towards researcher bias at times regardless of the end format. To disregard infographics or despise them versus black and white text and Excel tables doesn't make sense. It is a medium, and a powerful medium at that, but so is text, photography and videos, and they can all be used to objective ends or subjective ends, and none is ever pure and holy or automatically accurate (nor are Excel tables and the raw data used to inform the stylized infographics).

joubert 17y ago
Text may be black and white physically, but can express a wider range of nuance.