Why Megan Garber isn't credible (The Atlantic, racist tweets)
She offers "Where America's Racist Tweets Come From"  in which she hypes a "study" by "Floating Sheep" (!) mapping supposed racist tweets issued after Barack Obama's reelection. The study is flawed for two main reasons:
1. There were only 395 tweets total that matched the filters "Floating Sheep" ("FS") used.
2. What FS calls "racist" may just be attempts to troll or attempts at satire. Garber highlights three "racist" tweets, yet one is clearly from a satire account . The other is from someone who's retweeted a satire account and who appears to be somewhere between a troll and a satire . FS has not as far as I know published the tweets they used so it's not clear that their sample included Garber's tweets. However, FS admits they didn't attempt to determine whether a tweet was meant seriously or not. In fact, they say "most [tweets in their study] are derogatory in nature".  "Most"? What sort of study is this?
So much for the study: who knows how many tweets were satires, attempts to troll, or even congratulatory in nature?
As for FS, they admit their own biases, saying "we refuse to acknowledge the equivalence of the terms being used to describe President Obama and Mitt Romney".  In other words, some forms of racism are OK.
The study is bad, agenda-driven "science", and it's one that Megan Garber is promoting without any cavaets whatsoever.
 Ask yourself: is @QueenOfFlop ("Madonna") a satire account, yes or no?