,
11 tweets,
8 min read
Read on Twitter
Our new report on info quality in the 2018 election (w/@iBALyons, Montgomery, @andyguess & @JasonReifler) www-personal.umich.edu/~bnyhan/fake-n…
Key findings:
-exposure to fake news ↓ from 2016
-public disbelieves fake news but less when congenial
-Facebook ad exposure quite limited
[Thread]
Key findings:
-exposure to fake news ↓ from 2016
-public disbelieves fake news but less when congenial
-Facebook ad exposure quite limited
[Thread]
Using two different measures, we find that the percentage of American visiting fake news websites declined in October 2018 compared to October 2016. Evidence is mixed, however, on the share of people's news diets coming from fake news sites. www-personal.umich.edu/~bnyhan/fake-n… [2/7]
Unlike in fall 2016, we no longer observe Facebook views appearing differentially often just before fake news exposure www-personal.umich.edu/~bnyhan/fake-n… [3/7]
Despite the changes in fake news consumption and distribution that we observe, one pattern remains constant - consumption is still dominated by the Americans with the most extreme news diets on the right www-personal.umich.edu/~bnyhan/fake-n… [4/7]
How good are people at identifying fake news? Encouragingly, people rated headlines from false/hyperpartisan stories as less accurate than mainstream news. However, congenial headlines were rated as more accurate for every type of content. www-personal.umich.edu/~bnyhan/fake-n… [5/7]
In addition, higher levels of political knowledge did not consistently protect against misperception belief, especially among Republicans on the Soros/caravan myth. www-personal.umich.edu/~bnyhan/fake-n… [6/7]
Finally, with help from @ProPublica, we examined the reach of Facebook ads. Among users with eligible browsers who use FB, we found ad exposure was low (average=7), concentrated among liberals & conservatives, and included many non-candidate groups. www-personal.umich.edu/~bnyhan/fake-n… [7/7]
@ProPublica For more on the implications of these findings, see my accompanying @Medium column on why fears of fake news are overhyped
medium.com/s/reasonable-d… [8/7]
(We are also grateful for generous support from @DemocracyFund, which is not responsible for our findings or conclusions.)
medium.com/s/reasonable-d… [8/7]
(We are also grateful for generous support from @DemocracyFund, which is not responsible for our findings or conclusions.)
Partisanship in everything, anti-Trump spice company edition: newyorker.com/culture/annals… (a fascinating sidebar to our Facebook ads data)
