The truth behind filter bubbles: bursting some myths
1. THE TRUTH BEHIND FILTER
BUBBLES
RICHARD FLETCHER
GREEN TEMPLETON COLLEGE, UNIVERSITY OF OXFORD (22/01/2019)
2. BACKGROUND
People use services like Facebook, Twitter, Google, Apple News, etc. to get
news, and some of the news people see has been selected automatically by
algorithms.
Selection decisions are made by algorithms using data about our interests
and preference—and this could reinforce existing consumption patterns.
• Echo chambers: Where we are over-exposed to news we like or agree
with—distorting our perception of reality.
• Filter bubbles: Where news we dislike or disagree with is automatically
filtered out—narrowing what we know.
4. DIGITAL NEWS REPORT
World’s largest annual survey of news
audiences
38 markets, 5 continents (mostly Europe)
75,000 respondents
Polling mainly by YouGov, questionnaire by
us
www.digitalnewsreport.com
5. DIGITAL NEWS REPORT
What is people’s main source of news? (Newman et al., 2019)
46%
41%
7% 6%
0%
25%
50%
Online (inc. social media) TV Print Radio
Q4. You say you’ve used these sources of news in the last week , which would you say is your MAIN source of news? Base: All who used a source of news in the
last week. All markets = 73,527.
6. DIGITAL NEWS REPORT
What is people’s main source of news (by age)? (Newman et al., 2019)
67%
5% 5%
23%
61%
5%
6%
28%
53%
5%
7%
36%
43%
6% 7%
44%
30%
9%
7%
54%
0%
25%
50%
75%
Online Print Radio TV
18-24
25-34
35-44
45-54
55+
Q4. You say you’ve used these sources of news in the last week , which would you say is your MAIN source of news? Base: All who used a source of news in the
last week in all markets. 18-24 = 7,929, 25-34 = 12,836, 35-44 = 13,619, 45-54 = 13,038, 55+ = 26,105.
7. DIGITAL NEWS REPORT
What proportion use social media for news (since 2013)? (Newman et al., 2019)
0%
25%
50%
75%
2013 2014 2015 2016 2017 2018 2019
German
y
Austria
Spain
UK
USA
Q3. Which, if any, of the following have you used in the last week as a source of news? Base: Total sample in each country in each year ~= 2,000.
10. PERSONALIZATION
Two types of personalization (Zuiderveen Borgesius et al., 2016):
• Self-selected personalization
• The personalization people do themselves
• Selective exposure (Stroud, 2011; Iyengar & Hahn, 2009)
• Pre-selected personalization
• The personalization done for people (usually by algorithms)
• Filter bubbles (Pariser, 2011), echo chambers (Sunstein, 2017)
Very important not to compare the effect of pre-selected personalization
with a hypothetical world where people do not personalize their news
consumption through self-selected personalization.
12. INCIDENTAL EXPOSURE
What is the effect of social media use
on people’s news diets?
• Social media combines self-
selected personalization with
pre-selected personalization.
• Algorithms might hide news
from people who are not
interested in it
13. INCIDENTAL EXPOSURE
Compared news diets of people who
do not use social media at all with
those who use it but not for news
(Fletcher & Nielsen, 2018a)
UK, USA, Italy, Australia
Studied the effect on different
demographic groups and different
social networks
14. INCIDENTAL EXPOSURE
Social media boosts people’s news increasing diversity of
sources.
• People who use social media for other reasons are
incidentally exposed to news.
• The boost is stronger for young people are those
uninterested in news
• Effect stronger for YouTube and Twitter than for
Facebook
15. INCIDENTAL EXPOSURE
• Most people are not that interested in the news
• The web is a high-choice media environment, and its easy
to opt out of news (Prior, 2005)
• But social media “incidentally exposes” people to news–
shows people news even when they are not looking for it.
16. AUTOMATED SERENDIPITY
What about using search engines for
news?
• Different from social media because
people are using search engines to
intentionally find news.
• Will search engines only show right-
wing sources to right-wing users, and
vice-versa?
17. AUTOMATED SERENDIPITY
Compared news diets of people
who search for news with people
who do not (Fletcher & Nielsen,
2018b)
UK, USA, Germany, Spain
Studied news diets in terms of
diversity and balance
18. AUTOMATED SERENDIPITY
Automated serendipity diversifies people’s news use
• People who use search engines use more sources
• … are more likely to use both left- and right-
leaning sources
• … and have news diets that are more balanced
between left and right
19. AUTOMATED SERENDIPITY
“Users with different political leanings
from different states were recommended
very similar news, challenging the
assumption that algorithms necessarily
encourage echo chambers.” (Nechushtai &
Lewis 2019)
20. WORK IN PROGRESS
Can be backed up with UK web tracking data
(unpublished work in progress).
21. OTHER STUDIES
No/weak evidence of filter bubbles/echo chambers:
Dubois & Blank, 2018; Messing & Westwood, 2014;
Barbera et al., 2015; Fletcher & Nielsen, 2017, 2018
Mixed evidence:
Bakshy et al., 2015; Flaxman et al., 2016
Strong evidence: …
22. POLARIZATION
But there might be different
problems with social media
news use…
Some evidence that exposure to
diverse views on Twitter is
leading to the polarization of
political attitudes (Bail et al.,
2018)
25. FUNDING JOURNALISM
There might be more pressing problems…
“news itself has never been financially viable as
a market-based good [and] has always been
primarily financed by arrangements based on
income derived from sources other than selling
news to consumers” (Picard, 2014, p. 51)
28. NEWS ATTRIBUTION
Do people remember where they get their news? (Kalogeropoulos, Fletcher &
Neilsen 2019)
29. PAYING FOR ONLINE NEWS
Growth in the number of
paywalls (Simon & Graves 2019).
Only a minority pay for online
news (Newman et al., 2019).
Slight increases in online news
payment since 2014.
Nordic countries are doing quite
well.
UK and Germany not doing so
well.
30. PAYING FOR OTHER MEDIA
Millions of global paid subscribers (2018)
31. TRUST IN THE NEWS
In most countries fewer than half trust most news most of the time
(Newman et al., 2019)
32. TRUST IN THE NEWS
… and the proportion that trust most
news most of the time is falling in
many countries (Newman et al., 2019).
33. CONCLUSION
At the moment the evidence suggests that online news
use on search and social media is more diverse, but
this diversity might be polarizing.
In some ways this is the opposite of what the filter
bubble hypothesis predicted.
We should continue to critically examine the effects of
algorithmic selection on news use.
But we should not let this prevent us from properly
confronting the deeper causes of divisions in politics
and society (Bruns, 2019).
34. REFERENCES
• Bail, Christopher A., Lisa P. Argyle, Taylor W. Brown, John P. Bumpus, Haohen Chen, M. B. Fallin Hunzaker, Jaemin
Lee, Marcus Mann, Friedolin Merhout, and Alexander Volfovsky. 2018. ‘Exposure to Opposing Views on Social Media
Can Increase Political Polarization’. Proceedings of the National Academy of Sciences in the United States of America 0
(0): 1–6.
• Bakshy, Eytan, Solomon Messing, and Lada A. Adamic. 2015. ‘Exposure to Ideologically Diverse News and Opinion on
Facebook’. Science 348: 1130–32.
• Barberá, Pablo, John T. Jost, Jonathan Nagler, Joshua A. Tucker, and Richard Bonneau. 2015. ‘Tweeting from Left to
Right: Is Online Political Communication More Than An Echo Chamber’. Psychological Science 26: 1531–42.
• Bruns, Axel. 2019. Are Filter Bubbles Real? Cambridge: Polity.
• Dubois, Elizabeth, and Grant Blank. 2018. ‘The Echo Chamber Is Overstated: The Moderating Effect of Political Interest
and Diverse Media’. Information, Communication & Society 21 (5): 729–45.
• Flaxman, Seth, Sharad Goel, and Justin M. Rao. 2016. ‘Filter Bubbles, Echo Chambers, and Online News Consumption’.
Public Opinion Quarterly 80: 298–320.
• Fletcher, Richard, and Rasmus Kleis Nielsen. 2017. ‘Are News Audiences Increasingly Fragmented? A Cross‐National
Comparative Analysis of Cross‐Platform News Audience Fragmentation and Duplication’. Journal of Communication 67
(4): 476–98.
• ———. 2018a. ‘Are People Incidentally Exposed to News on Social Media? A Comparative Analysis’. New Media &
Society 20 (7): 2450–68.
• ———. 2018b. ‘Automated Serendipity: The Effect of Using Search Engines on the Diversity and Balance of News
Repertoires’. Digital Journalism 8 (6): 976–89.
35. REFERENCES
• Iyengar, Shanto, and Kyu S. Hahn. 2009. ‘Red Media, Blue Media: Evidence of Ideological Selectivity
in Media Use’. Journal of Communication 59 (1): 19–39.
• Kalogeropoulos, Antonis, Richard Fletcher, and Rasmus Kleis Nielsen. 2019. ‘News Brand Attribution
in Distributed Environments: Do People Know Where They Get Their News?’ New Media & Society 21
(3): 583–601.
• Messing, Solomon, and Sean J. Westwood. 2014. ‘Selective Exposure in the Age of Social Media:
Endorsements Trump Partisan Source Affiliation When Selecting News Online’. Communication
Research 41: 1042–63.
• Nechushtai, Efrat, and Seth C. Lewis. 2018. ‘What Kind of Gatekeepers Do We Want Machines to Be?
Filter Bubbles, Fragmentation, and the Normative Dimensions of Algorithmic Recommendations’.
Computers in Human Behavior 0 (0): 1–33.
• Newman, Nic, Richard Fletcher, Antonis Kalogeropoulos, David A. L. Levy, and Rasmus Kleis Nielsen.
2018. ‘Reuters Institute Digital News Report 2018’. Oxford: Reuters Institute for the Study of
Journalism.
• Newman, Nic, Richard Fletcher, Antonis Kalogeropoulos, and Rasmus Kleis Nielsen. 2019. ‘Reuters
Institute Digital News Report 2019’. Oxford: Reuters Institute for the Study of Journalism.
• Pariser, Eli. 2011. Filter Bubbles: What the Internet Is Hiding from You. London: Penguin.
36. REFERENCES
• Picard, Robert G. 2014. ‘State Support for News: Why Subsidies? Why Now? What Kinds?’ In State
Aid for Newspapers: Theories, Cases, Actions, edited by Paul Murschetz. Berlin: Springer.
• Prior, Markus. 2005. ‘News vs. Entertainment: How Increasing Media Choice Widens Gaps in Political
Knowledge and Turnout’. American Journal of Political Science 49: 577–92.
• Simon, Felix M., and Lucas Graves. 2019. ‘Pay Models for Online News in the US and Europe: 2019
Update’. Oxford: Reuters Institute for the Study of Journalism.
• Stroud, Natalie Jomini. 2011. Niche News: The Politics of News Choice. Oxford: Oxford University
Press.
• Sunstein, Cass R. 2017. #republic: Divided Democracy in the Age of Social Media. Princeton: Princeton
University Press.
• Thurman, Neil, and Richard Fletcher. 2018. ‘Are Newspapers Heading Toward Post-Print Obscurity? A
Case Study of The Independent’s Transition to Online-Only’. Digital Journalism 6 (8): 1003–17.
• Zuiderveen Borgesius, Frederick J., Damian Trilling, Judith Möller, Balázs Bodó, Claes H. de Vreese,
and Natali Helberger. 2016. ‘Should We Worry about Filter Bubbles?’ Internet Policy Review 5 (1): 1–
16.
37. THE TRUTH BEHIND FILTER
BUBBLES
RICHARD FLETCHER
GREEN TEMPLETON COLLEGE, UNIVERSITY OF OXFORD (22/01/2019)
Editor's Notes
Thank you for the introduction and for inviting me to give this lecture.
I’m very glad to be here Salzburg to speak to you all.
I’m going to try to describe how news use is changing in many countries across the world. What I really mean by this is how digital, online access is changing how we get news.
But I also want to leave you with my overall assessment of the situation.