4 ms·
I wish they hadn't chosen colour schemes that seem intended to bias the reading. In the first infographic, white characters are very low-contrast while brown ch
by rebuilder 6y ago
I wish they hadn't chosen colour schemes that seem intended to bias the reading. In the first infographic, white characters are very low-contrast while brown characters stand out. In the other, both seem to be about equal in terms of contrast.
That said, I'm not sure how to get this kind of presentation right, since the "white" population is not the same size as the "black" or "brown" population. They say a black person is almost four times as likely to be arrested for drug offenses as a white person. While I think their graphic seems a bit manipulative, at the same time it doesn't seem to capture the real awfulness of the situation. Non-Hispanic whites are something like 60% of the US population, while black people account for 13%, so just showing drug arrests by skin colour paints a very rosy picture of the actual situation.
My suggestion, normalize contrasts and weight the number of characters by demographics. It'd be pretty shocking and would avoid this odd whiff of dishonesty.
- dumbfoundded 6y agoThe first graphic is racially biased according to the arrest rates. The second graphic simply is based on overall population demographics. My intention was to be able to visually see the difference between consumers and arrests.