Key actor in this saga: People saw 70-90% Clinton win probability; didn't understand that 1-2% polling error would move that by 30-40%.
-
-
Replying to @zeynep
That's how everyone made pre-election decisions. That's the lens to understand then. Failure of infosec and data modeling shaped it all.
3 replies 16 retweets 63 likes -
Replying to @zeynep
zeynep tufekci Retweeted
There is a whole field on risk communication and information visualization that works on how to do this responsibly. https://twitter.com/briankgolden/status/807450442598912000 …
zeynep tufekci added,
This Tweet is unavailable.4 replies 9 retweets 57 likes -
Replying to @zeynep
zeynep tufekci Retweeted
Yes, 538 was the best, and tried to explain & was resisted. Many worse ones. Even 538 needed to communicate better. https://twitter.com/Mike_Knew/status/807454359961227268 …
zeynep tufekci added,
This Tweet is unavailable.3 replies 6 retweets 23 likes -
Replying to @zeynep
I think NYT's way of explaining it in terms of professional NFL kicker's chance of missing an x yard field goal was reasonable.
1 reply 0 retweets 0 likes
No, that's the model probability. What they needed to explain was what correlated error could mean.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.