So there were.... 3 mainstream media articles, out the many TENS OF THOUSANDS of articles published on the 2016 campaign, that embedded Russian troll tweets? I'm afraid that's not really helping your argument.https://twitter.com/zeynep/status/1075845985182441472 …
-
-
Replying to @NateSilver538
Yes, that's exactly my argument, right? Three troll tweets is all there is, and that plausibly swung a close election.
Let me know if you want to argue against a stronger strawperson and I'll get back to it.3 replies 2 retweets 119 likes -
Replying to @zeynep
Your argument is lacking in any sort of systematic evidence, at all. If you think impression counts are an imperfect measure -- and surely they are -- then come up with something better.
7 replies 2 retweets 38 likes -
Replying to @NateSilver538
The mere count and placement of DNC hack stories, alone, would be a pretty strong metric. It dominated media plus displaced other stuff. As with anything complex, it's a complex story but just that one is their operation—though obviously "the patient" had very weak immune system.
1 reply 3 retweets 35 likes -
Replying to @zeynep
You're -- deliberately? -- confusing two things. The DNC hack could well have been important. There's no evidence that Russian shitposts on Instagram mattered much, though.
5 replies 0 retweets 25 likes -
Replying to @NateSilver538
Not separable. The DNC hack was very much supported by massive operation on social media. It was almost all you'd see in some corners, mostly as screenshots with false claims that went viral. Plus, yep, mass media got played. Hence my call for looking at ecology, not percentages.
1 reply 2 retweets 63 likes -
Replying to @zeynep
I put "you have to look at the whole ecology" into my Google Academese translator and it came out as "I don't have any evidence".
17 replies 3 retweets 42 likes -
Replying to @NateSilver538
Okay! Keep looking at percent of tweets as a good metric for impact—and doctors should weigh viruses/bacteria as percent of bodymass to decide on impact and not look at how it impacts body/immune system (aka the whole ecology) . Good luck, to you and Dr. Google. ¯\_(ツ)_/¯
10 replies 6 retweets 122 likes -
Replying to @zeynep @NateSilver538
I'm open to these claims, but they ultimately need to generate testable hypotheses or we're not doing science. We have testable hypotheses about viruses and the immune system!
7 replies 4 retweets 71 likes -
Bookmarking this thread for PhD student methods training! We have a great set of faculty in the Science, Technology & Society program here at UM which largely embraces ways of knowing outside of testable hypotheses.
1 reply 0 retweets 4 likes
And the divide isn't quantitative vs qualitative. A lot of quantitative methods (in many fields, not just social science) do not fit into the "falsifiable/not" framework—especially for complex systems.
-
-
Yes, this thread has been really valuable for taking me through different perspectives on research, data, and analysis~thank you for it!
0 replies 0 retweets 1 likeThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Can I get some names of quantitative methods don't fit falsifiable/non-falsifiable" framework?
0 replies 0 retweets 0 likesThanks. Twitter will use this to make your timeline better. UndoUndo
-
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.