Rand rr4212 (1)

A RAND Initiative to Restore the Role of Facts and Analysis in Public Life
MICHAEL POLLARD I JENNIFER KAVANAGH
NEWSCONSUMPTION
PROFILES OF
Platform Choices, Perceptions of Reliability, and Partisanship

Limited Print and Electronic Distribution Rights
This document and trademark(s) contained herein are protected by law. This representation
of RAND intellectual property is provided for noncommercial use only. Unauthorized
posting of this publication online is prohibited. Permission is given to duplicate this
document for personal use only, as long as it is unaltered and complete. Permission is
required from RAND to reproduce, or reuse in another form, any of its research documents
for commercial use. For information on reprint and linking permissions, please visit
www.rand.org/pubs/permissions.
The RAND Corporation is a research organization that develops solutions to public
policy challenges to help make communities throughout the world safer and more secure,
healthier and more prosperous. RAND is nonprofit, nonpartisan, and committed to the
public interest.
RAND’s publications do not necessarily reflect the opinions of its research clients and sponsors.
Support RAND
Make a tax-deductible charitable contribution at
www.rand.org/giving/contribute
www.rand.org
Library of Congress Cataloging-in-Publication Data is available for this publication.
ISBN: 978-1-9774-0343-8
For more information on this publication, visit www.rand.org/t/RR4212
Published by the RAND Corporation, Santa Monica, Calif.
© Copyright 2019 RAND Corporation
R® is a registered trademark.
Cover: csvbbass9455/stock.adobe.com and BillionPhotos.com/stock.adobe.com
Cover design by Pete Soriano

iii
Preface
The information ecosystem is changing rapidly, with new forms of
media and news outlets and an array of different types of content. As
a result of these changes, the decisions made by news consumers have
become increasingly complex and varied. This report examines how
people access news, what their assessments are of the reliability of dif-
ferent news media platforms, and how their news consumption profiles
or other personal and demographic characteristics shape those con-
sumption decisions and perceptions of reliability.
The report is one of a series funded by internal RAND Corpo-
ration funds and published through RAND’s Office of the President
that focuses on the topic of Truth Decay, defined as the diminishing
role that facts and data play in today’s political and civil discourse. The
original report, Truth Decay: An Initial Exploration of the Diminishing
Role of Facts and Analysis in American Public Life by Jennifer Kavanagh
and Michael D. Rich, was published in January 2018 and laid out a
research agenda for studying and developing solutions to the Truth
Decay challenge. This report is part of that initiative.
RAND Ventures
The RAND Corporation is a research organization that develops solu-
tions to public policy challenges to help make communities through-
out the world safer and more secure, healthier and more prosperous.
RAND is nonprofit, nonpartisan, and committed to the public interest.

iv Profiles of News Consumption
RAND Ventures is a vehicle for investing in policy solutions.
Philanthropic contributions support our ability to take the long view,
tackle tough and often-controversial topics, and share our findings
in innovative and compelling ways. RAND’s research findings and
recommendations are based on data and evidence, and therefore do
not necessarily reflect the policy preferences or interests of its clients,
donors, or supporters.
Funding for this venture was provided by gifts from RAND
supporters and income from operations. For more information about
RAND Ventures, visit www.rand.org/giving/ventures.

v
Contents
Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
Figures and Tables.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii
Abbreviations.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix
CHAPTER ONE
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Objective of This Report.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Study Approach and Data .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Contribution of This Report. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Organization of This Report. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
CHAPTER TWO
News Consumption and Attitudes Toward Media: Review of Past
Work. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
What We Know About News Consumption. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
How Attitudes Toward Media and News Vary.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
CHAPTER THREE
News That People Consume: News Consumption Profiles. . . . . . . . . . . . . . . 25
How People Get Their News. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Attitudes About News Reliability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Trust in Institutions and Media Consumption.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

vi Profiles of News Consumption
CHAPTER FOUR
Discussion.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Four Motivating Questions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Overall Findings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Factors Associated with News Consumption. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Implications for Truth Decay.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
APPENDIXES
A. Ranked Data Factor Analysis Procedure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
B. Study Descriptive Statistics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
C. Trust in Institutions and News Consumption.. . . . . . . . . . . . . . . . . . . . . . . . . 77
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

vii
Figures and Tables
Figures
3.1. Respondents’ Most Used News Source. . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.2. Respondent’s Top Two Most-Used News Sources. . . . . . . . . . . . . . . 27
3.3. Respondents’ Top Two Perceived Most-Reliable News
Sources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
A.1. Scree Plot and Variance Explained by Factor Analysis,
Suggesting Four Factors Should Be Retained. . . . . . . . . . . . . . . . . . . . . 72
Tables
3.1. Characteristics Associated with News Consumption
Profile Factors (OLS Coefficients).. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.2 . Summary of Media Consumption Profile Characteristics:
Demographics Only. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.3. Summary of Media Consumption Profile Characteristics
(Considering Partisanship). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.4. Characteristics Associated with Perceptions of News
Reliability (Relative Risk Ratios from Multinomial
Logistic Regression) .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.5. Summary of Associations Between Consumption Profiles
and Perceptions of News Reliability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.6. Perceived Most Reliable News Platform, Compared with
Broadcast Television (Relative Risk Ratios). . . . . . . . . . . . . . . . . . . . . . 40

viii Profiles of News Consumption
and Most Reliable Platform: Demographic and Political
Characteristics.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.8. Perceived Most Reliable News Platform, Compared with
Broadcast Television, Including News Consumption Profile
(Relative Risk Ratios). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
and Perceived Most Reliable Platform.. . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.10. Primarily Gets News from One of Their Top Two
Perceived Most-Reliable Platforms (Odds Ratios; vs.
Does Not).. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.11. Summary of Variables Linked to an Individual’s
Likelihood of Getting News from at Least One of Their
Two Perceived Most-Reliable Platforms.. . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.12. How Often Respondent Seeks Out News with Different
Views (“Always or Almost Always” or “Never or Almost
Never” vs. “Sometimes” or “Infrequently”; Relative Risk
Ratios).. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.13. Summary of Characteristics Associated with Likelihood
of Seeking Out News with Different Views . . . . . . . . . . . . . . . . . . . . . . 53
4.1. Summary of Relationships Between News Consumption
Profiles and Other Characteristics and Behaviors. . . . . . . . . . . . . . . . 59
4.2. Summary of Relationships Between Perceptions of Overall
News Reliability and Other Characteristics and Behaviors. . . . 60
4.3. Summary of Results Related to Political Partisanship. . . . . . . . . . . 61
A.1. EFA Estimated Loadings and Scoring Coefficients. . . . . . . . . . . . . . 73
B.1. Study Variable Descriptive Statistics (Unweighted). . . . . . . . . . . . . . 76
C.1. Most Reliable News Platform, Compared with Broadcast
Television, Including Trust Measures (Relative Risk Ratios).. . . 79
C.2. Characteristics Associated with News Consumption Profile
Factors, Including Trust Measures (OLS Coefficients). . . . . . . . . 80

ix
Summary
In this report, we explore how U.S. media consumers obtain news. We
examine the types of dissemination platforms (different forms of media
delivery; e.g., online, television, print) that consumers rely on and the
relationship between consumers’ news consumption profiles (the specific
mix of news platforms consumed; e.g., print and broadcast television,
radio, social media) and their overall perceptions of media reliability and
the reliability of specific news platforms. We also explore the extent to
which people use platforms that they have identified as reliable and their
willingness to seek out news from viewpoints that are different from their
own, in light of interest in and concern about echo chambers or filter
bubbles. Finally, we consider the extent to which demographic charac-
teristics and political partisanship are linked to each of these aspects of
news consumption.1 Using a nationally representative survey of English-
speaking individuals 21 and older, we address several questions:
• How do Americans get their news?
• How are news consumption choices linked to demographic or
political characteristics?
• Do news consumers believe the reliability of news has changed, and
which news platforms do they believe to be more or less reliable?
• How is the perceived reliability of news associated with news con-
sumption choices?
1 Here and elsewhere in this report, we define partisanship as an individual’s tendency to
support one particular group, cause, or viewpoint over another. We measure this through
self-rated political ideology and voting behaviors.

x Profiles of News Consumption
In pursuing these questions, we aim to advance the discussion
about the relationships among media consumption, trust in media,
and Truth Decay—the diminishing reliance on facts, data, and analy-
sis in American public life.
Key Findings
News Consumption Profiles
The analysis identified four news consumption profiles differentiated
by clusters of how frequently individuals relied on different combina-
tions of several news platforms:2 print and broadcast television plat-
forms, online platforms (e.g., newspaper websites, such as nyt.com),
radio, and social media (e.g., Twitter and Facebook) and in-person
communication. Print and broadcast television were commonly used
together, as were social media and in-person communication. Note
that although we measured cable television use, these data did not help
differentiate profiles of news consumption because cable television use
was ubiquitous as an important news platform for a large portion of
individuals. An individual’s news consumption patterns are reflected
in how highly they score on all four of these profiles.
Identifying these news consumption profiles and their links
to other aspects of news consumption represents a key contribu-
tion of this report. These profiles are useful for several reasons. Most
importantly, they provide a more holistic picture of news consumers,
allowing us to identify which types of sources are often used together
and the specific demographic characteristics associated with specific
patterns of news consumption. These consumption profiles can be fur-
2 News consumption profiles were determined from a survey conducted in February and
March 2018 of 2,543 English-speaking members of the RAND American Life Panel (ALP),
which consists of 6,000 U.S. respondents ages 18 and older who regularly take surveys over the
internet. The survey asked a randomly selected subset of the panel about their perceptions of
the reliability of various sources of news and about how they most frequently got their national
and international news. We used factor analysis to identify common patterns in information
access within the sample. Respondents reported their relative level of use of print, broadcast
television, cable television, online news platforms, radio, social media, and in-person commu-
nication to access the news. This approach was data-driven in that the groupings were deter-
mined from patterns and combinations in the data rather than being predefined.

Summary xi
ther linked to perceptions about news and news consumption behav-
iors above and beyond demographic influences.
In comparing the news consumption profiles, we determined that
each profile has a distinct set of demographic characteristics. Among
them are the following:
• People whose primary news sources are social media and in-person
contacts are generally younger and female, and they tend to have
less education than a college degree and lower household incomes.
• People whose primary news sources are print publications and
broadcast television tend to be significantly older, and they are less
likely to be married.
• People whose primary news source is radio are significantly more
likely to be male, less likely to be retired, and more likely to have
a college degree.
• People whose primary news sources are online platforms are signif-
icantly younger, more likely to be male and have a college degree
and higher income, and less likely to be black.
Taken together, older individuals reported relying primarily
on more-traditional platforms (print/broadcast television) for news;
younger people reported relying more heavily on social media and in-
person communication or on online platforms. Women were more likely
than men to report getting their news through social means (social media
or in-person contacts). Non-Hispanic white respondents were generally
more likely than others to report relying heavily on the social media/
in-person and online platforms for obtaining their news.
In parallel with identifying news consumption profiles, we
used the survey data to explore how respondents participated in the
Hillary Clinton–Donald Trump presidential election in 2016 and how
their news consumption (as measured here) was linked to their involve-
ment in the race (e.g., via social media). Our analysis indicates that par-
tisanship and voting behavior appear to be linked to news consump-
tion behavior. This manifested itself in several ways:
• Voters who backed someone other than Clinton or Trump in
2016 were more likely than those who voted for Clinton to score
highly on the social media/in-person profile.

xii Profiles of News Consumption
• Respondents who rated their political ideology as more liberal
were somewhat more likely to be in the online group than those
who rated themselves as more conservative.
• Respondents who reported voting for Trump were significantly
less likely to score highly on the print/broadcast and online news
consumption profiles than those who voted for Clinton.
Reliability of News Media
We also asked respondents to share their perceptions regarding whether
the news media have become more or less reliable than in the past and
to rank types of news platforms by perceived reliability.3
Overall, our survey respondents had mixed perceptions about the
general reliability of the news. Although 44 percent reported that they
believed “the news is as reliable now as in the past,” nearly the same
amount—41 percent—reported a belief that the news has become
less reliable, while a minority (15 percent) said that they believed that
the news is more reliable now. People who were retired, had higher
incomes, or voted for someone other than Clinton were significantly
more likely to believe the news is less reliable now. Black or Hispanic
individuals, those with less education than a college degree, and those
who voted in 2016 (versus nonvoters) all were more likely to report that
the news is more reliable now. There was also an association between
news consumption profiles and perceptions of reliability: People who
relied more heavily on online, radio, and social media/in-person plat-
forms to obtain news were less likely to say that news is more reliable
now than in the past. Perceptions of how the reliability of news has
changed over time were linked to consumption behaviors; those who
said the news is more reliable now were more likely to report that they
“always or almost always” seek out differing viewpoints in their news
than those who did not perceive a change; those who believed the news
3 We use the term reliability instead of trust because we are more interested in perceptions
of media information than of institutions and because we are more interested in the extent
to which respondents view various sources as dependable rather than trustworthy. However,
our analyses do establish that reliability and trust are correlated.

Summary xiii
is less reliable now than in the past were less likely to rely heavily on
news from platforms they themselves rated as most reliable.
Perceptions of Most-Reliable Platforms
Broadcast and cable television were perceived to be the most-reliable
platforms by the largest number of people in our survey. The reliability
of other platforms was ranked as follows: print, online news sites, radio,
social media, and in-person communication. People generally reported
getting news from sources they rated as the most reliable—particularly
those who relied more heavily on print and broadcast television. The
exceptions were people who reported getting most of their news from
social media and in person despite those platforms’ low reliability rank-
ings and people who believed that the news is less reliable now than in
the past.
Search for Alternative Viewpoints
An additional aspect of news consumption behavior that we asked
about was “how often do you seek out sources of news that you know
will offer views that are different from your own?” This question was
motivated by recent attention to the negative effects of echo chambers
and filter bubbles, defined by Merriam-Webster as “an environment and
especially an online environment in which people are exposed only to
opinions and information that conform to their existing beliefs.”4 One
in five respondents (20 percent) reported that they “always or almost
always” seek out different views; most people (54 percent) said that
they “sometimes” do; 17 percent said that they “infrequently” seek out
differing sources; and 9 percent said they “never or almost never” do.
Married people were particularly less likely to report that they
“always or almost always” seek out opposing news sources (versus “some-
times” or “infrequently”); black respondents and those of other races
were significantly more likely than others to report “never or almost
never” doing so. Respondents with higher education and higher incomes
were significantly less likely than others to respond that they “never or
almost never” seek out opposing news sources. Furthermore, people who
scored highly in the online news consumption profile were significantly
4 Merriam-Webster, “Filter Bubble,” webpage, undated.

xiv Profiles of News Consumption
more likely to report “always or almost always” seeking out sources of
news that offer differing viewpoints and significantly less likely to report
that they “never or almost never” seek out differing views. Individuals
who said that they perceived any change (either increase or decrease)
in reliability of news now compared with the past were less likely to
report “never or almost never” seeking out differing views in the news
than those who did not believe there had been a change (although only
those who believed the news is more reliable now were more likely than
those who did not believe there had been a change in reliability to report
“always or almost always” seeking out differing views).
Looking at these results in terms of partisanship and voting
behavior patterns also could provide interesting insights. People who
rated their general political ideology as more liberal were more likely to
report that they “never or almost never” sought out sources that they
knew would offer views that are different from their own, as were those
who voted for Clinton (compared with those who voted for Trump).
Taken together, this suggests that, compared with other voters or non-
voters, those who voted for Clinton are less skeptical of more-traditional
news delivery platforms, tend to favor print as their source for news, are
much less likely to believe that the reliability of news has declined over
time, and are less likely to rate online or in-person news as the most-
reliable platforms. Those who voted for Clinton were also less likely
than other voters to report that they “always or almost always” seek out
news sources that offer views different from their own. Furthermore,
those who did not vote in the 2016 presidential election were less likely
than others to believe there had been an improvement in the reliability
of news over time and, similar to Trump voters, were more likely to rate
social media as the most reliable news platform.
Implications
Our results indicate that political ideology and partisanship are associ-
ated with how people consume news and their attitudes toward news
but that demographic factors—such as age, gender, level of education,

Summary xv
and employment status—are also important to news consumption
choices. We consider several implications of our findings.
Consumption Choices and Social Media
Our results speak to the rising concerns about the increasing reliance of
consumers on social media as a source of news. In this report, we pro-
vide insight into the degree to which consumers rely on social media
for news and in what ways. About 30 percent of respondents reported
relying on social media as their primary or secondary source of news,
ranking it somewhere in the middle of news platforms in terms of use:
It is neither the most used nor least used news platform. However, our
data indicate that people rarely use social media as their sole source of
news, and that those 30 percent of consumers who do rely heavily on
social media for news are likely also using other news sources. These
findings are consistent with research on media complementarity, the
notion that different media platforms might serve different needs for
information users.
News Consumption and Demographic Characteristics
Our results suggest that people use news media in different ways—
relying on different types of news delivery platforms—and hold vary-
ing perceptions of reliability regarding those platforms and the news
overall. Political partisanship explains some of these differences but
certainly not all of them. Such factors as voting behaviors, age, gender,
education, and marital status all play significant roles in explaining
news consumption behaviors. As we note throughout this report, dif-
ferent news consumption profiles reflect aspects of an individual’s
demographics, such as living situation.
Our results reflect choices between platforms, not specific outlets,
but these insights still have implications for efforts aimed at improv-
ing and supporting how people obtain and evaluate the news. Specifi-
cally, our analysis suggests that attention to reducing barriers to access,
such as time limitations, might be as important as considering media
literacy education as a way to provide information users with the skills
they need to navigate a more complex online environment.

xvi Profiles of News Consumption
Perceptions of Media Reliability
Finally, we considered perceptions of media reliability and trust in the
media. Our analysis confirms that many people are skeptical of the
reliability of news overall, but it also shows that a small minority (par-
ticularly black and Hispanic individuals and those without a college
degree) think that reliability is increasing. This underscores the impor-
tance of considering disaggregated data when studying media con-
sumption and trust in the media. Attitudes about reliability also vary
across different types of news platforms, with broadcast and cable news
being considered the most-reliable sources, as previously discussed.
Our analysis also shows that attitudes toward the media are
linked to consumption. For most users, reliability matters. News con-
sumers tend to rely most heavily on the sources that they perceive as
most reliable—though there are some exceptions, such as those who
rely on other people and social media for news; some married people;
and, unexpectedly, those who report believing that news is less reliable
now than in the past. Skepticism about the reliability of the news does
not necessarily increase an individual’s likelihood of turning to news
platforms they deem to be most reliable—in fact, we see the opposite.
That said, for all consumers, choices about media consumption
are likely linked to individual characteristics, preferences, and behav-
iors. As a result, attempts to alter or improve relationships between
media outlets and consumers are likely to require attention not just to
the supply side of the information ecosystem (how media institutions
operate and function) but also to the demand side and the ways in
which people access news. Future work that explores news consump-
tion in this more holistic way will be important as the informational
ecosystem grows increasingly diverse and increasingly complex.

xvii
Acknowledgments
We would like to thank Michael D. Rich for his support and guidance
throughout this process. We also thank David Grant, Karen Edwards,
and Julie Newell for their assistance in developing and programming
this survey. Wenjing Huang provided expert assistance on our data
analysis. Shawn Smith provided valuable assistance as we prepared the
final document. The authors also thank Gordon Lee for his assistance
with our revisions.
As part of the RAND Corporation’s quality assurance process,
this report had three anonymous reviewers, one a RAND colleague
and two external experts. We are grateful for their helpful feedback.
Krishna Kumar and Rebecca Kilburn provided comments that allowed
us to improve the document. Any errors are the authors’ own.

xix
Abbreviations
ABC American Broadcasting Company
ALP American Life Panel
CBS Columbia Broadcasting Service
CNN Cable News Network
EFA exploratory factor analysis
MSNBC Microsoft/National Broadcasting Company
NBC National Broadcasting Company
OLS ordinary least squares
PEPS Presidential Election Panel Survey

1
CHAPTER ONE
Introduction
The news, in whatever form it comes, provides voters in a democracy
with crucial information about domestic and international politi-
cal events and policy issues that can inform their policy preferences
and, ultimately, their voting decisions. Americans have an increasing
number of news choices, in terms of both numbers of outlets and media
platforms.1 News can be found on more-traditional platforms (such as
newspapers, radio, and television) or on newer platforms (such as digi-
tal journalism or social media posts). However, these options are varied
in terms of content, tone, and intended audience. Social media, for
instance, provides news in 280-character blocks (e.g., Twitter), whereas
newspapers and some radio programs can do deep dives into specific
topics. Similarly, the audience for cable television is very different from
the audience for long-form print journalism.
In this increasingly diverse media ecosystem, news consumers’
decisions can be based on any number of factors, such as availability,
cost, time to consume, quality, or perceived reliability of a given outlet.
Given this diversity, knowing something about what types of news
people consume, how that news consumption is linked with demo-
graphic and political characteristics, and how perceived reliability of
the consumed information factors into consumption patterns all can
provide useful insights about both the overall market for news and the
1 News platforms refers to different forms of media delivery—for example, websites, televi-
sion, and print publications. We use the terms sources and outlets interchangeably to refer to
the different components of each platform.

2 Profiles of News Consumption
roles that such factors as partisanship, education, and race all play in
news platform and outlet choices.
Objective of This Report
Using a nationally representative survey of English-speaking individu-
als ages 21 and older, this report addresses four main questions:
1. How do Americans get their news?
2. How are news consumption choices linked to demographic or
3. Do news consumers believe the reliability of news has changed,
and which news platforms do they believe to be more or less
reliable?
4. How is the perceived reliability of news associated with news
consumption choices, such as platform choices and seeking out
alternative viewpoints in the news?
Study Approach and Data
Our approaches to each of the study’s four questions build on previ-
ous work (which we review in detail in subsequent sections) in sev-
eral ways. First, we use rank-ordered data on a greater variety of news
media platforms than has been used previously—consisting of print,
broadcast television, cable television, radio, internet, social media, and
in-person communication—combined with statistical methods to con-
struct novel news consumption profiles characterizing the combinations
through which news is accessed. Throughout this report, we describe
how these news consumption profiles are linked to a variety of socio-
demographic characteristics, individual behaviors, and attitudes.
Second, although perceptions of the relative reliability of news
platforms have been examined previously, our study reflects what is, to
our knowledge, the greatest variety of platforms considered simultane-
ously and one of the first attempts to develop such detailed descrip-

Introduction 3
tive assessments of the platforms used by individuals to obtain news.
Although prior work has considered perceptions of several platforms at
once, none has examined the full complement of different platforms
featured in our analysis. We explore associations between perceived
reliability and consumption decisions across platforms in a way that
builds on previous analyses.
Finally, we examine what our analytical results suggest about
Truth Decay, the phenomenon defined in a previous RAND Corpo-
ration report as the diminishing reliance on facts, data, and analysis
in American public life.2 As described in that report, the increas-
ing diversity of the information ecosystem has both positive and
negative implications. On one hand, having more options to choose
from certainly empowers news consumers. On the other hand, an
increase in different types and sources of information that might
vary in their accuracy, their vulnerability to disinformation, and the
extent to which they clearly distinguish between fact and opinion
might make choices of what to consume more difficult. Our survey
cannot explore all of these issues, but, in Chapter Four, we consider
the implications of our survey results and analysis for Truth Decay—
specifically whether people get their news from platforms they rate
to be the most reliable, and whether they seek out diversity in the
viewpoints of their news.
Our analysis and discussion rely on a survey conducted in Feb-
ruary and March 2018 of 2,543 English-speaking members of the
RAND American Life Panel (ALP) that asked about their perceptions
of the reliability of various sources of news and about how they most
frequently got their national and international news.3 The sample for
2 Jennifer Kavanagh and Michael D. Rich, Truth Decay: An Initial Exploration of the
Diminishing Role of Facts and Analysis in American Public Life, Santa Monica, Calif.: RAND
Corporation, RR-2314-RC, 2018. Previous RAND work defines Truth Decay as comprising
four specific trends: an increasing disagreement about objective facts and analytical interpre-
tations of data; blurring of the line between opinion and fact; the increasing relative volume
of opinion (and its resulting influence) compared with facts; and declining trust in institu-
tions, particularly those previously viewed as trusted sources of factual information.
3 The ALP Omnibus survey did not have a Spanish language version, which is a limitation of
our analysis. Note that the 2016 American Community Survey, conducted by the U.S. Census

this survey was randomly drawn from the larger ALP sample.4 The
ALP is a panel survey that consists of 6,000 U.S. respondents ages 18
and older who regularly take surveys over the internet.5
Since January 2006, the ALP has fielded more than 500 surveys
on such topics as financial decisionmaking, the effect of political events
on self-reported well-being, joint retirement decisions, health decision-
making, Social Security knowledge and expectations, measurement of
health utility, and voting preference in the presidential election. Data
from all surveys are made publicly available to more than 600 clients
and registered users from numerous institutions.
More than 100 research papers have been published using
RAND ALP data. As with other surveys, we weight responses to
ensure that our results are representative of the U.S. English-
speaking population, matching to the U.S. Census Bureau’s Current
Population Survey.6
Because the ALP is a panel survey, we can construct a rich pro-
file of individuals using past surveys—dating back to 2006 for some
participants. This additional information about survey participants is
one of the primary contributions of this study. We can connect news
Bureau, indicates that 91.5 percent of Americans speak English only or “speak English very
well.” Thus, results should not necessarily be interpreted as reflecting the 8.5 percent of the
population that speaks English “less than ‘very well.’” See U.S. Census, “Language Spoken at
Home by Ability to Speak English for the Population 5 Years and over,” webpage, 2017.
4 The survey was designed for a maximum number of 2,500 respondents because of bud-
geting constraints. A random sample of 3,357 were invited to take the survey in anticipation
of a one-month field period; 2,570 had completed it after three weeks, and the survey was
closed at that time.
5 ALP respondents are originally sampled by either random digit dial (landline and cell
phone) or address; individuals cannot otherwise volunteer to participate. A further advan-
tage over most other internet panels is that the respondents to the ALP need not have internet
access when they are initially recruited (RAND provides laptops and internet subscriptions
if needed), reducing an important source of bias. The panel is thus based on a probability
sample of the U.S. population.
6 Complete technical details of the ALP, its recruitment, retention, and weighting pro-
cedures are available in Michael Pollard and Matthew D. Baird, The RAND American Life
Panel: Technical Description, Santa Monica, Calif.: RAND Corporation, RR-1651, 2017; and
at RAND Corporation, “Welcome to the ALP Data Pages,” webpage, undated.

Introduction 5
consumption today with not only political behaviors in past elections
but also a variety of demographic characteristics in a way that existing
work has not been able to do. This broader perspective provides us with
some additional insights.
Other Surveys
As noted in later chapters, we have linked information where it was pos-
sible to do so from the news reliability survey to two other ALP surveys:
the 2016 RAND Presidential Election Panel Survey (PEPS) and a 2018
survey on trust in institutions.7 The PEPS data contain information on
political partisanship and voting behavior during the 2016 presidential
election, which supplements our questions about perceptions of broader
sources of news. The trust in institutions data provide additional infor-
mation on respondents’ levels of trust in various types of media, which
is used here to validate our measures of reliability. In Chapter Three, we
provide information on the number of respondents and specific ques-
tions used in the surveys; we also discuss limitations of the analysis.
Contribution of This Report
In this report, we seek to build on past literature by providing a new
perspective on questions that have been asked before and addressing
some gaps in past literature, building on the following key insights
from this past work. First, individuals typically use more than one type
of media to get their news. Each news consumer, then, might have a
unique basket of media platforms, and that mix might vary depend-
ing on the specific topic or types of sources available. Second, choices
about which types of media to consume might be driven by individual
characteristics, preferences, and market factors. Third, news consump-
tion can be conceptualized as part of an individual’s lifestyle, driven by
7 Katherine Carman, “Well Being 496: Trust in Government,” RAND American Life
Panel, webpage, April 2018; Michael Pollard and Joshua Mendelsohn, Methodology of the
2016 RAND Presidential Election Panel Survey (PEPS), Santa Monica, Calif.: RAND Corpo-
ration, RR-1460-RC/UCLA, 2016; and Michael Pollard and Joshua Mendelsohn, “RAND
Kicks Off 2016 Presidential Election Panel Survey,” The RAND Blog, January 27, 2016.

specific motivations and filling specific needs. (We review past work
on news consumption decisions in detail in Chapter Two, summariz-
ing several key insights and providing citations to the related work.)
Past research also considers the issue of trust in the media in an
era of declining faith in the news overall, focusing on perceptions of
the relative credibility of different news platforms, who is most likely to
trust different outlets, and, in some cases, factors that lead to increased
or diminished trust. Political partisanship, in particular, has been
highlighted as having a strong role in determining overall views of
media credibility for decades, with higher conservatism and Republi-
can partisanship predicting greater mistrust. Age and education have
also been identified as predictors of differing levels of trust in specific
news platforms. Prior work suggests that both television and online
news platforms are vying for the most trusted position. Finally, past
research suggests that perceived credibility might not be all that closely
related to consumption decisions. (We discuss this work in more detail
in Chapter Two.)
This report advances past research in a few ways. First, building
on the evidence that individuals use multiple news platforms in com-
plementary ways, we use survey data to explore how people combine
different types of information across an expanded array of options,
identifying four news consumption profiles on which individuals rely
to varying degrees. There have been previous efforts to do this,8 and
our analysis expands on this past work, providing richer detail and
identifying the mix of different platforms and the demographic charac-
teristics associated with each consumption type while also considering
the degree of use of other types of platforms. Given that there is interest
in how people consume news and concern about the quality of infor-
mation available, a clearer and updated picture of the mix of sources
used by different groups of people could be very informative and repre-
sents a key contribution of our report. A deeper analysis of who is most
likely to consume which mix of different platforms is also valuable.
8 Kristin Van Damme, Cédric Courtois, Karel Verbrugge, and Lieven De Marez, “What’s
APPening to News? A Mixed-Method Audience-Centred Study on Mobile News Consump-
tion,” Mobile Media & Communication, Vol. 3, No. 2, 2015.

Introduction 7
Second, although there has been significant research into both
how people consume news and the extent to which people trust
news, there has been less research that looks at the intersection of
those topics. That is, do individuals with different news consump-
tion profiles have different attitudes toward those media platforms
and the reliability of that information? Do they have different con-
sumption behaviors? The answers to these questions provide insights
into the characteristics of different types of news consumers and can
inform our understanding of when and why people do and do not
trust media organizations. In addition to considering how news con-
sumption profiles might relate to perceptions of media reliability, we
also explore how perceptions of media reliability are linked to news
consumption by comparing respondent reports about reliability with
their actual consumption decisions.
Third, rather than asking about trust in media, we ask about
reliability, seeking to understand the extent to which various types of
media are dependable or produce dependable information. A compari-
son of these same respondents’ reports about trust in media and their
responses about reliability explores how the two concepts are related
and ultimately indicates they are viewed similarly by news consumers.
(We discuss this analytical choice in more detail in Chapter Three.)
Organization of This Report
In Chapter Two, we discuss what we know from past research about
news consumption decisions, how individuals interact with news,
and their attitudes toward news media. In Chapter Three, we discuss
our analysis and results. We start with an assessment of the broad
platforms from which people obtain their news and then identify
four news consumption profiles using this information. We consider
the question of perceived news reliability—how people rank different
platforms and how perceived reliability is linked to news consumption
choices. We then consider how willing people are to seek out alter-
native viewpoints and assess how that willingness varies across news
consumption profile, partisanship, and other demographic charac-

teristics. In the final chapter, we explore what the insights gained
from these analyses suggest more generally about news consumption
and how they might inform our understanding of Truth Decay. The
three appendixes at the end of the report feature details of the statis-
tical procedure used to identify news consumption profiles, descrip-
tive statistics for the variables used in the report, and a supplemen-
tary analysis exploring the links between perceived reliability of news
platforms and explicit trust in particular news platforms.

9
CHAPTER TWO
News Consumption and Attitudes Toward Media:
Review of Past Work
There has already been significant research examining how people
get news, the factors that shape their news consumption decisions,
how news consumption fits into their lifestyles, and their general
attitudes toward both media and news. In this chapter, we outline
the key findings and insights of this past work, identify some exist-
ing limitations, and then discuss how our approach addresses some
of these limitations.
What We Know About News Consumption
Research on patterns in news consumption explores not only the
sources of news that people rely on but also the individual and envi-
ronmental factors that influence news consumption decisions and how
news consumption choices change over time as new types of media and
more-diverse sources emerge.
Number of Sources: Displacement or Complementarity
Past research is clear that people typically use several different types of
media and that this mix of platforms features both traditional media
(print, radio, television) and new media (online news, social media).
However, debate continues regarding whether different forms of news
should be considered as complements or substitutes. A number of stud-

ies advance an argument of complementarity,1 which argues that con-
sumers are most likely to seek out a mix of different types (platforms) of
news, perhaps to meet different needs, and behave no differently with
new media than they do with traditional media.2 Those who argue
for complementarity suggest that consumers integrate new information
sources into their existing media diet to complement their media con-
sumption but might not eliminate older sources of information from
their media consumption portfolio.
Exploring patterns of use, Van Damme et al. find some support
for this argument, identifying three types of users: those who consume
an extensive and diverse diet of news, those who consume primarily
traditional sources of media with limited online content, and those
who consume news less frequently and, when they do, rely on digital
sources.3 Similarly, Ahlers finds that 51 percent of those who get their
news through online sources also relied on other sources.4
Ahlers’ information is dated, but the pattern itself (of multichan-
nel news consumption) remains relevant. Ahlers proposes a “consumer
centric” model of news consumption, in which different users rely on
different combinations of different platforms that are based on the
users’ own preferences and lifestyles.5 Dutta-Bergman also finds evi-
1 Mohan J. Dutta-Bergman, “Complementarity in Consumption of News Types Across
Traditional and New Media,” Journal of Broadcasting & Electronic Media, Vol. 48, No. 1,
2004; Eugenia Mitchelstein and Pablo J. Boczkowski, “Online News Consumption
Research: An Assessment of Past Work and an Agenda for the Future,” New Media & Soci-
ety, Vol. 12, No. 7, 2010.
2 Joey Ka-Ching Chan and Louis Leung, “Lifestyles, Reliance on Traditional News Media
and Online News Adoption,” New Media & Society, Vol. 7, No. 3, 2005; Steve Coffey and
Horst Stipp, “The Interactions Between Computer and Television Usage,” Journal of Adver-
tising Research, Vol. 37, No. 2, 1997; Andrew J. Flanagin and Miriam J. Metzger, “Internet
Use in the Contemporary Media Environment,” Human Communication Research, Vol. 27,
No. 1, 2001, pp. 153–181; Joseph M. Kayany and Paul Yelsma, “Displacement Effects of
Online Media in the Socio-Technical Contexts of Households,” Journal of Broadcasting &
Electronic Media, Vol. 44, No. 2, 2000; and Mitchelstein and Boczkowski, 2010.
3 Van Damme et al., 2015.
4 Douglas Ahlers, “News Consumption and the New Electronic Media,” Harvard Interna-
tional Journal of Press/Politics, Vol. 11, No. 1, 2006.
5 Ahlers, 2006.

News Consumption and Attitudes Toward Media: Review of Past Work 11
dence to support this complementarity in consumption arguing that
individuals rely on multiple different types of media to collect infor-
mation on topics that are of interest to them.6 Finally, Gaskins and
Jerit specifically look at the replacement of traditional outlets by online
sources; although they find some evidence of replacement, they note
that the extent of replacement remains somewhat limited.7 Of course,
this might have changed in the years since the original study was
conducted.
An alternative view is held by those who argue for displacement,
which suggests that as new sources emerge, they replace or crowd out
older and more-traditional sources of information. In this view, infor-
mation consumers have only so much bandwidth for news, so attention
devoted to one type of media will crowd out other media platforms. As
evidence of this displacement theory, proponents point to the generally
declining subscribership of major newspapers, especially among young
readers, and the increasing number and diversity of those who con-
sume news online.8 Empirical tests that seek to validate the displace-
ment theory find mixed results: Although displacement does occur for
some forms of media and for some types of users, it does not appear to
occur universally.9 While the extent of displacement may be debated,
there is clear evidence that it is occurring. A 2018 study, for instance,
6 Dutta-Bergman, 2004.
7 Benjamin Gaskins and Jennifer Jerit, “Internet News: Is It a Replacement for Traditional
Media Outlets?” International Journal of Press/Politics, Vol. 17, No. 2, 2012, pp. 190–213.
8 Matthew Gentzkow, “Valuing New Goods in a Model with Complementarity: Online
Newspapers,”AmericanEconomicReview,Vol.97,No.3,2007;BarbaraK.KayeandThomasJ.
Johnson, “From Here to Obscurity? Media Substitution Theory and Traditional Media in an
Online World,” Journal of the American Society for Information Science and Technology, Vol. 54,
No. 3, 2003; and Carolyn Lin, Michael B. Salwen, Bruce Garrison, and Paul D. Driscoll,
“Online News as a Functional Substitute for Offline News,” in Michael B. Salwen, Bruce
Garrison, and Paul D. Driscoll, eds., Online News and the Public, Mahwah, N.J.: Lawrence
Erlbaum, 2005.
9 Thomas F. Baldwin and Marianne Barrett, “Uses and Values for News on Cable Televi-
sion,” Journal of Broadcasting and Electronic Media, Vol. 36, No. 2, 1992; Stan J. Liebowitz
and Alejandro Zentner, “Clash of the Titans: Does Internet Use Reduce Television Viewing?”
Review of Economics and Statistics, Vol. 94, No. 1, 2012; and Laura Hazard Owen, “Here’s Who
Gets News from TV: The Elderly, Pew Finds (Again),” Neiman Lab, January 8, 2018.

found that young people shifted heavily into digital news consump-
tion over the period from 1976 to 2016, suggesting a crowding out of
traditional sources of media by social media for younger consumers.10
Survey data provide a mix of support for both the displacement
and complementarity debates. On the side of complementarity, recent
surveys by the Pew Research Center show that although an increasing
number of Americans rely heavily on digital sources for their news—
whether that is the online websites of newspapers, digital journalism
sites, or social media—most continue to rely on traditional forms of
media (such as newspapers, radio, or television) at the same time.11
Other patterns of media use appear to suggest some evidence for
a displacement effect. While the number of Americans who seek news
on social media and online has been increasing, the percentage rely-
ing on television and newspapers has experienced a long-term decline.
Fifty-seven percent of Americans relied on television news in 2016; only
50 percent reported using television for news in 2017. Readership of print
newspapers has fallen significantly, dropping to 31 million on weekdays
(from a high of about 63.3 million in 1984) and 34 million on Sundays
in 2017 (from a high of about 62.5 million in 1993).12 As television and
print journalism audiences decline, there has been a sharp increase in
10 Jean M. Twenge, Gabrielle N. Martin, and Brian H. Spitzberg, “Trends in US Ado-
lescents’ Media Use, 1976–2016: The Rise of Digital Media, the Decline of TV, and the
(Near) Demise of Print,” Psychology of Popular Media Culture, August 2018. See also Oscar
Westlund and Mathias A. Färdigh, “Accessing the News in an Age of Mobile Media: Tracing
Displacing and Complementary Effects of Mobile News on Newspapers and Online News,”
Mobile Media & Communication, Vol. 3, No. 1, 2015.
11 Amy Mitchell, Jeffrey Gottfried, Elisa Shearer, and Kristine Lu, How Americans Encoun-
ter, Recall and Act Upon Digital News, Washington, D.C.: Pew Research Center, February 9,
2017.
12 Michael Barthel, “Newspapers Fact Sheet,” webpage, Pew Research Center, June 13,
2018. An exception to the declining trends in more-conventional news sources is radio. The
audience for radio broadcasting remains high—in 2017, 90 percent of Americans listened
to some type of radio broadcast (both news and other programming). About 10 percent of
radio audiences at any given point in the day are listening to news or talk radio. The online
radio and podcast audience also has been growing steadily even in the face of new forms of
media, running counter to some displacement arguments. In 2018, almost 50 percent of
Americans reported having listened to a podcast at some point in their lives and 26 percent
reported having listened to a podcast in the past month. This represents an increase of about

the use of online news sources—websites of newspapers, streaming news
programs, and social media.13 Importantly, recent surveys on media
use show that the use of online sources of news is increasing among all
demographic groups, including older Americans. In 2017, for instance,
55 percent of respondents older than 50 also reported using social media
as a primary source of news.14 These trends could be more consistent
with a displacement argument, although the data provided do not explic-
itly reveal how much switching between media platforms is occurring
(displacement) or what portion of observed trends is driven by new users
entering the online market versus conventional media consumers opting
out of news consumption more generally.
Past work exploring displacement and complementarity is useful
for thinking about the relationships between different types of media
and how individuals might optimize and select different platforms
and sources to fill their information needs. Although debates about
displacement and complementarity continue, past work suggests that
individuals rely on multiple sources; even as an increasing number of
consumers rely on online information and social media, use of more-
traditional sources has not disappeared. Instead of thinking of con-
sumers as “social media users” or “newspaper readers,” it might be more
accurate to think about each consumer as relying on a mix of different
types of news for different types of subjects and existing in the spaces
between news platforms. We will bring this set of expectations to our
quantitative analysis.
Factors That Shape News Consumption Decisions
As the number of types of sources grows (along with the number of
specific outlets within each of those types), an increasingly important
and complex question focuses on which factors influence an individual’s
choices about which information sources to use. Past research identifies
2.5 times over the past decade. Elisa Shearer, “Audio and Podcasting Fact Sheet,” webpage,
Pew Research Center, July 12, 2018.
13 Barthel, 2018.
14 Elisa Shearer and Jeffrey Gottfried, “News Use Across Social Media Platforms 2017,”
webpage, Pew Research Center, September 7, 2017.

a variety of factors—such as demographic, social, and structural ones—
that might influence an individual’s news consumption attitudes.15 Look-
ing first at demographic characteristics, such factors as age, race, gender,
and level of education have all been found to significantly influence
decisions about information consumption.16 For example, past research
shows that men are more likely than women to engage with online news,
particularly on social media. Younger consumers also appear more likely
to rely on social media for news, although this might be a function of the
amount of time they spend on the platform.17 Consumers with higher
incomes and more education also are more likely to be readers of online
news.18 In contrast, consumers of print newspapers tend to be older, more
educated, more mobile, often urban, and more likely to be male, black,
or unmarried.19 Finally, it is worth noting that choices about news con-
sumption also might reflect the interaction of individual characteristics
and desired content. Although there is certainly overlap, different forms
of media package news differently and so might appeal to different types
of consumers.20 These findings are also generally consistent with the Pew
survey data already described.
15 Alan M. Rubin, “Uses and Gratifications: An Evolving Perspective of Media Effects,”
in Robin L. Nabi and Mary Beth Oliver, eds., The SAGE Handbook of Media Processes and
Effects, Thousand Oaks, Calif.: SAGE Publications, 2009; and Harsh Taneja, James G.
Webster, Edward C. Malthouse, and Thomas B. Ksiazek, “Media Consumption Across Plat-
forms: Identifying User-Defined Repertoires,” New Media and Society, Vol. 14, No. 6, 2012.
16 Angela M. Lee, “News Audiences Revisited: Theorizing the Link Between Audience
Motivations and News Consumption,” Journal of Broadcasting & Electronic Media, Vol. 57,
No. 3, 2013; and Taneja et al., 2012.
17 Carroll J. Glynn, Michael E. Huge, and Lindsay H. Hoffman, “All the News That’s Fit
to Post: A Profile of News Use on Social Networking Sites,” Computers in Human Behavior,
Vol. 28, No. 1, 2012; and Twenge, Martin, and Spitzberg, 2018.
18 Guido H. Stempel and Thomas Hargrove, “Mass Media Audiences in a Changing Media
Environment,” Journalism & Mass Communication Quarterly, Vol. 73, No. 3, September 1996.
19 Leo Bogart, “The Public’s Use and Perception of Newspapers,” Public Opinion Quarterly,
Vol. 48, No. 4, January 1, 1984.
20 Thomas E. Ruggiero, “Uses and Gratifications Theory in the 21st Century,” Mass Com-
munication & Society, Vol. 3, No. 1, 2001.

Other factors appear less significant to influencing choices of dif-
ferent information platforms. For example, there is strong evidence that
partisanship and ideology affect choices of individual news outlets, but
there is less evidence that ideological motivations are a significant driv-
ing factor in the choice of platform—newspaper, online sources, tele-
vision, or radio—except through those platforms’ relationships with
desired media content and demographic characteristics. Work on selec-
tive exposure shows that the search for pro-attitudinal information
(that which confirms one’s beliefs or attitudes) guides decisions about
which types of content to consume—for example, decisions on choos-
ing media outlets, television programs, or specific journalists.21 How-
ever, there is less evidence that partisan identification shapes choices
regarding modes of media consumption once age and other demo-
graphic characteristics are accounted for.
Beyond these individual factors, previous research suggests that
news consumption choices also are shaped by the structure of news
markets. This research indicates that the concentration and diversity
of available programming (local newspapers, local television channels,
state and national news options) determine the options available and
ultimately the news consumption behaviors of users. In other words,
information supply likely plays as significant a role in information con-
sumption decisions as do individual characteristics.22 In emphasizing
market structure, past work also underscores the importance of geog-
raphy. Specifically, past research shows significant local and regional
patterns in news consumption that extend beyond geographic cluster-
ing of individual characteristics and preferences.23 Thus, it is not only
21 R. Kelly Garrett and Natalie Jomini Stroud, “Partisan Paths to Exposure Diversity: Dif-
ferences in Pro- and Counterattitudinal News Consumption,” Journal of Communication,
Vol. 64, No. 4, 2014; William Hart, Dolores Albarracín, Alice H. Eagly, Inge Brechan, Mat-
thew J. Lindberg, and Lisa Merrill, “Feeling Validated Versus Being Correct: A Meta-Analysis
of Selective Exposure to Information,” Psychological Bulletin, Vol. 135, No. 4, 2009; and Nata-
lie J. Stroud, Niche News: The Politics of News Choice, New York: Oxford University Press, 2011.
22 Scott L. Althaus, Anne M. Cizmar, and James G. Gimpel, “Media Supply, Audience
Demand, and the Geography of News Consumption in the United States,” Political Com-
munication, Vol. 26, No. 3, 2009.
23 Althaus, Cizmar, and Gimpel, 2009.

individual preferences but also local and regional news markets and
even an individual’s own access to that market that determine the mix
of sources that each individual relies on for news. Furthermore, local
and regional market structure also relates directly to the amount of
exposure an individual has to the news.24
Finally, in addition to individual and market characteristics,
the lifestyle, general preferences, and motivations of news consumers
appear to shape news consumption choices and patterns. Past research
on the relationship between news consumption, motivation, and life-
style suggests that we cannot consider news consumption choices in a
vacuum but must explore the ways that news consumption interacts
with an individual’s other responsibilities, preferences, and motiva-
tions. Peters, for example, argues that news consumption is an activity
that is attached to specific places and times, and so becomes part of
an individual’s everyday activity.25 He notes that “the spaces of news
consumption matter, and matter significantly, for how audiences expe-
rience journalism.”26 Different types of news and journalism might
have different types of relationships with space and time. For example,
research suggests that online media might be used in a greater diver-
sity of places and in shorter periods of time than other forms of media,
which might more often be consumed in the same places and same
times.27
Other theories about news consumption decisions suggest that
these choices are based on individual information needs and the search
for gratification.28 In other words, individuals will rely on informa-
24 Althaus, Cizmar, and Gimpel, 2009.
25 Chris Peters, “Journalism to Go,” Journalism Studies, Vol. 13, Nos. 5–6, 2012.
26 Peters, 2012, p. 696.
27 Taneja et al., 2012; and Cornelia Wolf and Anna Schnauber, “News Consumption in the
Mobile Era,” Digital Journalism, Vol. 3, No. 5, 2015.
28 Jay G. Blumler and Elihu Katz, The Uses of Mass Communications, Beverly Hills, Calif.:
Sage Publications, 1974; Edgar Huang, “The Causes of Youths’ Low News Consumption
and Strategies for Making Youths Happy News Consumers,” Convergence: The Interna-
tional Journal of Research into New Media Technologies, Vol. 15, No. 1, 2009; Marina Krcmar
and Yuliya Strizhakova, “Uses and Gratifications as Media Choice,” in Tilo Hartmann,
ed., Media Choice: A Theoretical and Empirical Overview, New York: Routledge, 2009; Lee,

tion sources providing them with information they need or want or
affording them the most gratification. There are many possible moti-
vations that might drive individuals in their information consump-
tion decisions. Previous research suggests that users might seek infor-
mation, answers to questions, a diversion from boredom, comfort or
confirmation, entertainment, or social connection through their infor-
mation consumption.29 A framework by Lee,30 for example, proposes
four main sets of motivations guiding news consumption decisions:
(1) entertainment;31 (2) ideologically driven choices, rooted in the search
for confirmatory information;32 (3) socially motivated consumption;33
and (4) information seeking.34 Evaluating the prevalence of these four
motivations, Lee finds that people are most likely to be motivated by
the search for information and least likely to seek out news for reasons
related to ideology and opinion.35 However, it is likely that these two
motivations are intertwined for many people. Lee also notes that news
2013; and Carolyn Lin, “Modeling the Gratification-Seeking Process of Television View-
ing,” Human Communication Research, Vol. 20, No. 2, 1993.
29 Flanagin and Metzger, 2001; Lee, 2013; and Zizi Papacharissi and Alan M. Rubin, “Pre-
dictors of Internet Use,” Journal of Broadcasting & Electronic Media, Vol. 44, No. 2, 2000.
30 Lee, 2013.
31 Jay D. Hmielowski, R. Lance Holbert, and Jayeon Lee, “Predicting the Consumption of
Political TV Satire: Affinity for Political Humor, The Daily Show, and The Colbert Report,”
Communication Monographs, Vol. 78, No. 1, 2011.
32 R. Kelly Garrett, “Echo Chambers Online? Politically Motivated Selective Exposure
Among Internet News Users,” Journal of Computer-Mediated Communication, Vol. 14, No.
2, 2009; and Shanto Iyengar and Kyu S. Hahn, “Red Media, Blue Media: Evidence of Ideo-
logical Selectivity in Media Use,” Journal of Communication, Vol. 5, 2009.
33 Vivian B. Martin, “Attending the News: A Grounded Theory About a Daily Regimen,”
Journalism, Vol. 9, No. 1, 2008; Michele Weldon, Everyman News: The Changing American
Front Page, Columbia, Mo.: University of Missouri Press, 2008.
34 Matthias R. Hastall, “Information Utility as Determinant of Media Choices,” in Tilo
Hartmann, ed., Media Choice: A Theoretical and Empirical Overview, New York: Routledge,
2009; and Pamela J. Shoemaker, “Hardwired for News: Using Biological and Cultural Evo-
lution to Explain Surveillance Function,” Journal of Communication, Vol. 46, No. 3, 1996.
35 Lee, 2013.

consumption is motivated by multiple and overlapping motivations
rather than just one.
Another body of related research has focused on identifying
media “repertoires,” or multiplatform sets of news sources that media
users construct to meet their many different needs and preferences.
Repertoires might be multiple outlets, channels, forms of media, and
even means of delivery (e.g., radio via car stereo or streaming).36 The
concept of repertoires underscores the notion that individuals rely on
many sources of media simultaneously and that they are able to update
their media diet as media options become more diverse. We will return
to the concept of media repertoires in the next chapter.
The diversity of motivations becomes more important as news
consumers have access to an increasing diversity of types of informa-
tion, with important implications for levels of political knowledge
across the electorate. Prior shows that as content becomes more diverse,
those seeking detailed political information and those seeking enter-
tainment are each able to find the information that they are most inter-
ested in.37 As this sorting process occurs, those with political inter-
est become increasingly versed in political affairs; those without this
interest fall further and further behind in political knowledge. From
the perspective of understanding and studying news consumption pat-
terns, this research suggests that not only do individuals get informa-
tion through different types of platforms, they also likely get different
types of information and acquire different types of knowledge.38
How Attitudes Toward Media and News Vary
Just as individuals vary in their choices of modes of information con-
sumption, so do they vary significantly in their attitudes toward the
36 Taneja et al., 2012.
37 Markus Prior, “News vs. Entertainment: How Increasing Media Choice Widens Gaps
in Political Knowledge and Turnout,” American Journal of Political Science, Vol. 49, No. 3,
July 2005.
38 Prior, 2005.

news media, particularly their trust in media institutions and journal-
ists and their beliefs about the credibility or reliability of the informa-
tion provided by these information sources.
As noted, trust in media in the United States tends to be low
in aggregate terms. Audience trust in news media has been in a long
decline from a relative high in the 1970s.39 In fact, a 2018 Pew survey
suggests that American respondents consider the media to be signifi-
cantly less fair and less accurate than do respondents in other coun-
tries. Only 47 percent of Americans believe that the news media report
issues fairly; only 56 percent believe the media report issues accurately.
Compare this with Canada, where 73 and 78 percent of respondents,
respectively, hold these attitudes toward media’s fairness and accura-
cy.40 Americans want an independent and unbiased media and strongly
believe that this is important for democracy, but less than one-half can
name a source they consider objective.41 This could mean that there are
few objective sources available, that many consumers simply perceive
more sources to be biased, or that individuals tend to remember and
fixate on those that are generally not objective.
There are many reasons why trust in media appears to be low.
High-profile mistakes and concern about false information online and
in print and television journalism might be one reason. For example,
more than 50 percent of consumers say they believe that online news
sources report false information and 70 percent report feeling that
social media companies are not doing enough to stop the spread of
false information.42 A majority also say they believe that media compa-
39 Stephen Earl, Staci L. Rhine, and Richard S. Flickinger, “Assessing Americans’ Opinions
About the News Media’s Fairness in 1996 and 1998,” Political Communication, Vol. 18,
No. 2, 2001; and David A. Jones, “Why Americans Don’t Trust the Media: A Preliminary
Analysis,” Harvard International Journal of Press/Politics, Vol. 9, No. 2, 2004.
40 Amy Mitchell, Katie Simmons, Katerina Eva Matsa, and Laura Silver, “Publics Globally
Want Unbiased News Coverage, but Are Divided on Whether Their News Media Deliver,”
Pew Research Center, January 11, 2018.
41 Knight Foundation and Trust, Media and Democracy, “10 Reasons Why American Trust
in the Media Is at an All-Time Low,” Medium, January 15, 2018.
42 Christine Schmidt, “Planted Stories? Fake News as Editorial Decisions? Trump or CNN?
A Poll Examines the Public’s Trust of Mainstream Media,” Neiman Lab, April 3, 2018.

nies play a role in the problem not only by printing false information
but also because of the stories they choose to print.43 About 70 percent
of news consumers say publication of false information is intentional
sometimes or often, although this number is significantly higher for
Republicans (92 percent) than Democrats (53 percent).44 An increasing
number of sources might also be eroding trust in news and confusing
consumers seeking fact-based information.45
Attitudes toward the media are heavily affected by political atti-
tudes and partisanship in the United States. Consistently across sur-
veys, Republicans tend to be less trusting of the media and to rate
media coverage more negatively than Democrats, who view the media
more positively in general.46 More than one-half of Democrats view
the news media favorably; almost 70 percent of Republicans have an
unfavorable view.47 A recent study of trust in the media conducted by
the Knight Foundation in cooperation with Gallup found that news
consumers significantly reduce their perception of the trustworthi-
ness of news on learning that it comes from a source on the other side
of the political aisle (e.g., Fox News for Democrats or MSNBC for
Republicans).48 Past research also has found that the link between par-
tisanship and trust in media might reflect the effect of perceived bias;
individuals who perceive bias in media content tend to report lower
levels of trust in media as a result.49 Importantly, this partisan divide is
not new to this decade. Research on trust in the media has consistently
43 Schmidt, 2018.
44 Sara Fischer, “92% of Republicans Think Media Intentionally Reports Fake News,”
Axios, June 27, 2018.
45 Knight Foundation and Trust, Media and Democracy, 2018.
46 Albert C. Gunther, “Biased Press or Biased Public?” Public Opinion Quarterly, Vol. 56,
No. 2, 1992; Jones, 2004; and Tien-Tsun Lee, “Why They Don’t Trust the Media: An
Examination of Factors Predicting Trust,” American Behavioral Scientist, Vol. 54, No. 1,
2010.
47 Knight Foundation and Trust, Media and Democracy, 2018.
48 Knight Foundation, An Online Experimental Platform to Assess Trust in the Media, Miami,
Fla., July 18, 2018.
49 Lee, 2010.

found that those with conservative political views tend to have lower
trust in the media than other respondents.50 More generally, previous
work has found that trust in government and in the media are often
closely related: Individuals who have low trust in government also tend
to have lower trust in the media.51
Individuals rate the credibility of specific methods or platforms of
news provision differently, although methods of assessing these rank-
ings vary. In the earliest studies, newspapers were rated as more cred-
ible than television or radio.52 But beginning in the 1960s and lasting
through the end of the century, television steadily gained as the most
credible source, followed by newspapers and radio.53
With the growth in internet-based news dissemination, there
have been inconsistent reports of the perceived credibility of online
news sources: Some studies identify online news as being considered
more credible than television or newspapers;54 others report it to be
viewed as less credible.55 Some of this inconsistency might reflect the
fact that individuals appear to judge the credibility of different plat-
forms using different criteria. For instance, one study found that news-
50 Jones, 2004.
51 Stephen Earl Bennett, Staci L. Rhine, Richard S. Flickinger, and Linda L. M. Bennett,
“‘Video Malaise’ Revisited: Public Trust in the Media and Government,” Harvard Interna-
tional Journal of Press/Politics, Vol. 4, No. 4, 1999; Jones, 2004.
52 Matthias Kohring and Jörg Matthes, “Trust in News Media: Development and Valida-
tion of a Multidimensional Scale,” Communication Research, Vol. 34, No. 2, 2007.
53 Richard F. Carter and Bradley S. Greenberg, “Newspaper or Television: Which Do You
Believe?” Journalism Quarterly, Vol. 42, No. 1, 1965; Cecilie Gaziano and Kristin McGrath,
“Measuring the Concept of Credibility,” Journalism Quarterly, Vol. 63, No. 3, 1986; Kohring
and Matthes, 2007; and Bruce H. Westley and Werner J. Severin, “Some Correlates of Media
Credibility,” Journalism Quarterly, Vol. 41, No. 3, 1964.
54 Thomas J. Johnson and Barbara K. Kaye, “Cruising Is Believing? Comparing Internet
and Traditional Sources on Media Credibility Measures,” Journalism & Mass Communica-
tion Quarterly, Vol. 75, No. 2, 1998; and Jenn Burleson Mackay and Wilson Lowrey, “The
Credibility Divide: Reader Trust of Online Newspapers and Blogs,” Journal of Media Sociol-
ogy, Vol. 3, Nos. 1–4, 2011.
55 Spiro Kiousis, “Public Trust of Mistrust? Perceptions of Media Credibility in the Infor-
mation Age,” Mass Communication & Society, Vol. 4, No. 4, 2001; and Pew Research Center,
One-in-Ten Voters Online for Campaign ‘96, Washington, D.C., December 16, 1996.

papers were judged as an institution, whereas television news sources
are often judged based on the credibility of individual reporters who
use images and other evidence to convince viewers of their reliability.56
Focus has tended to be on the relative credibility of various news
platforms or how perceptions of credibility or trust in types of news
media relate to the use of those types of media. But a handful of previ-
ous studies explicitly focused on identifying background factors that
are associated with perceptions of the relative credibility of specific
news platforms. Bucy noted that younger, college-educated individu-
als said they believed television and online news sources were credible
overall more often than older individuals did, although older individu-
als rated online news as more credible than television (whereas younger
people rated television as more credible).57
Research has shown that perceptions of overall news media credi-
bility are not necessarily correlated with overall frequency of news con-
sumption.58 Past work does suggest, however, that perceived credibil-
ity might be associated with news consumption decisions—for choices
between platforms and between outlets.59 Although there are likely
exceptions, people tend to choose to use sources of news they trust over
sources they distrust.60 Looking specifically at cable news, for example,
56 John Newhagen and Clifford Nass, “Differential Criteria for Evaluating Credibility of
Newspapers and TV News,” Journalism Quarterly, Vol. 66, No. 2, 1989.
57 Erik P. Bucy, “Media Credibility Reconsidered: Synergy Effects Between On-Air and
Online News,” Journalism and Mass Communication Quarterly, Vol. 80, No. 2, 2003.
58 Wayne Wanta and Yu-Wei Hu, “The Effects of Credibility, Reliance, and Exposure on
Media Agenda-Setting: A Path Analysis Model,” Journalism Quarterly, Vol. 71, No. 1, 1994.
59 Ronald J. Faber, Stephen D. Reese, and Leslie H. Steeves, “Spending Time with the News
Media: The Relationship Between Reliance and Use,” Journal of Broadcasting and Electronic
Media, Vol. 29, No. 4, 1985; and Wanta and Hu, 1994.
60 Robert P. Hawkins, Suzanne Pingree, Jacqueline Hitchon, Bradley W. Gorham, Prathna
Kannaovakun, Eileen Gilligan, Barry Radler, Gudbjorg H. Kolbeins, and Toni Schmidt,
“Predicting Selection and Activity in Television Genre Viewing,” Media Psychology, Vol. 3,
No. 3, 2001; Nikolaus G. Jakob, “No Alternatives? The Relationship Between Perceived
Media Dependency, Use of Alternative Information Sources, and General Trust in Mass
Media,” International Journal of Communication, Vol. 4, 2010; Natalie Jomini Stroud and Jae
Kook Lee, “Perceptions of Cable News Credibility,” Mass Communication and Society, Vol.
16, No. 1, 2013; Yariv Tsfati and Joseph N. Cappella, “Do People Watch What They Do Not

Stroud and Lee show that perceived credibility of specific outlets is
an important factor in explaining the relationship between political
attitudes and cable news consumption choices.61 One question that is
less fully explored by past work, however, is how perceived credibility
interacts with demographic and political characteristics to shape cross-
platform trade-offs for news consumers in an increasingly diverse news
environment. We explore this question and others in our analyses in
the next chapter.
Trust? Exploring the Association Between News Media Skepticism and Exposure,” Commu-
nication Research, Vol. 30, No. 5, 2003; and Lawrence R. Wheeless, “The Effects of Attitude,
Credibility, and Homophily on Selective Exposure to Information,” Speech Monographs, Vol.
41, 1974, pp. 329–338.
61 Stroud and Lee, 2013.

25
CHAPTER THREE
News That People Consume: News Consumption
Profiles
As noted in the previous chapter, individuals vary in the number and
types of news sources that they consume. Some people might read hours
of news articles per day, others might listen to a single podcast, and still
others might rely on social media or conversations with friends. This
chapter directly explores the ways in which people obtain their news
and derives news consumption profiles that describe how heavily indi-
viduals rely on each of four typical avenues through which people get
their news. We also investigate how political partisanship—in terms
of both ideology and voting behavior—is linked to news consump-
tion and attitudes about the news. Our findings are generally consis-
tent with previous studies but offer additional nuance into how news
consumption and reliance on certain news platforms might reflect not
only such basic demographic characteristics as age and race but also
such broader contextual factors as marital and employment status.
How People Get Their News
We first asked people to tell us how they got most of their national and
international news, ranking the top four platforms in the order they
used them. We offered them several major media categories: print pub-
lications (e.g., newspapers and news magazines), broadcast television
(e.g., ABC, NBC, CBS), cable television (e.g., CNN, Fox, MSNBC),
radio news programs (including streaming; e.g., Sirius), online news
and newspaper websites (e.g., The Bill, Breitbart, Huffington Post,
nytimes.com), social media (e.g., Facebook, Twitter), and in-person

communication (e.g., friends, family, personal networks at such places
as work or church).1
The most common response for the primary news platform was
broadcast television, closely followed by online sources (Figure 3.1).
Cable television and social media were the second most-common
platforms, with radio, print, and in-person communication being less
common as primary platforms. When the first and second most-used
platforms were combined, the pattern generally remained the same
(Figure 3.2). The in-person platform was the least common choice
among respondents choosing their most used source of news, but it
was a popular second choice, which moved it up in the rankings to the
point where it was ultimately listed among the top two platforms in
our survey, above print or radio.
The data show that user preference for print journalism is less
common than for television, digital and online journalism, and social
1 Although comparing differences within these broad platforms would be helpful
(e.g., newspapers versus magazines within print, or even specific outlets or sources of print
news), the broader-channel comparison is more suitable for an exploratory study such as this.
Figure 3.1
Respondents’ Most Used News Source
0
5
10
15
20
25
30
35
40
45
50
Print Broadcast Cable Radio Online Social
media
In person
Percentagelistingsourceasfirstchoice
Source of news
6
24
19
8
23
16
4

News That People Consume: News Consumption Profiles 27
media as the primary source of news. On the other hand, despite con-
cern about Americans relying solely on social media for news,2 our data
show that only about 15 percent rely on it as a first choice and 14 per-
cent select it as a second choice. Finally, the survey indicates that about
one in five Americans relies on other people (e.g., friends, family, social
networks) as their first or second choice for news. This reinforces the
role that social networks play in spreading news.
We can push the data further to get more-detailed patterns of the ways
that people consume news. To better understand these patterns, we
used a statistical technique known as factor analysis to identify four
underlying common combinations of ranked news platforms used by
the people in our survey and to determine how strongly each of these
2 Katerina Eva Matsa and Elisa Shearer, “News Use Across Social Media Platforms 2018,”
webpage, Pew Research Center, September 10, 2018.
Figure 3.2
Respondent’s Top Two Most-Used News Sources
41
31
19
38
28
20
Percentagelistingsource
asfirstorsecondchoice
0
5
10
15
20
25
30
35
40
45
50
media
In person
Source of news
16

underlying combinations fits each respondent.3 News consumption
profiles, as we define them here, are similar to the media repertoires
described in Chapter Two, but our approach is distinct from this past
work both in our use of ranked data and in our focus at the platform
level (rather than the channel or outlet level).4
Identification of the four factors and how individuals scored on
each of them were determined based on the rank order that individuals
provided for the top four news platforms they used. Individuals scored
higher on factors that matched their own ranking patterns and lower
on factors that did not match. For example, someone who reported
that his or her most commonly used platforms (in order) were online,
print news, broadcast television, and radio would, in the factor anal-
ysis, score highest on online, followed by print/broadcast and radio,
with the social media/in-person category having the lowest score. The
factor analysis process allows us to use these different scores to sort
respondents into groups with similar news consumption use habits,
which we call news consumption profiles.
Our analysis identified four news consumption profiles compris-
ing the best description of the data: social media and in-person contact,5
print and broadcast television, radio, and online. Note that this indicates
that social media and in-person sources were commonly closely ranked,
as were print and broadcast television, leading those platform combina-
tions to be combined as single news consumption profiles.
A closer look at the data provides additional insight into these
groups. Although cable television is a relatively common method of
3 Broadly, factor analysis is a statistical technique that is used to reduce data to a smaller
set of summary variables (factors) and to explore their underlying structure. We initially
used exploratory factor analysis (EFA) to derive the number of underlying factors, and then
used confirmatory factor analysis (CFA) to fit the hypothesized factor structure to the data
(respondents). Factor scores for each dimension of the consumption profile have a mean of 0
and a standard deviation of 1 for the entire sample, which means individuals can have posi-
tive or negative values on each dimension. Complete details of the factor analysis procedures
are provided in Appendix A.
4 Taneja et al., 2012.
5 This grouping could also be described as “peer-to-peer news,” which is an appropriate way
to think of these platforms. We retain “social media and in-person” for clarity.

obtaining news, it was not identified as a significant consumption pro-
file factor in the final or even preliminary factor analysis. This suggests
that although cable news consumption is relatively ubiquitous, it does
not provide a distinct news consumption signature on its own in our
data. Broadly, an individual’s consumption profile factor score places
him or her most strongly in one category, but he or she also might have
characteristics of other categories. Although each individual tends to
score highly on one of the consumption profile dimensions, he or she
might also use other news platforms frequently; these sources are likely
to play a smaller role in that person’s overall news consumption, and
accordingly are given lower scores on those dimensions. If platforms
were not highly used or not used at all by individuals, factor scores for
those profile dimensions are negative. Note that this also means that
because cable television does not appear as one of the consumption
profile dimensions, for people who do rely heavily and nearly exclu-
sively on cable television, factor scores for all consumption profile fac-
tors are negative, indicative of the poor match between those profiles
and the respondent’s cable news consumption pattern.
The news consumption profiles exemplify the concept of media
repertoires used by individuals to combine multiple types of news
sources simultaneously to various degrees. The concepts of comple-
mentarity and displacement are less directly relevant but still provide a
way of thinking about how people develop their own media repertoires
and how sources within an individual’s repertoire might interact.
Demographic Characteristics of News Consumption Profiles
We found that each of our four news consumption profiles was also
demographically distinct. Demographic characteristics of respondents
in each news consumption profile are described in Tables 3.1 and 3.2.
Table 3.1 presents the results of a series of ordinary least squares (OLS)
regressions predicting each respondent’s factor score: Model 1 uses a
set of sociodemographic variables (age, sex [male/female], race, level of
education, household income, marital status, presence of children in
the household, and retirement status); Model 2 also considers partisan-
ship variables. Table 3.2 presents a verbal interpretation of the results
of Model 1 in Table 3.1. Descriptive statistics (means, proportions, and

30ProfilesofNewsConsumption
Table 3.1
Characteristics Associated with News Consumption Profile Factors (OLS Coefficients)
Characteristic
Social Media/
In-Person Print/Broadcast Radio Online
Model 1 Model 2 Model 1 Model 2 Model 1 Model 2 Model 1 Model 2
Age –0.019*** –0.020*** 0.002 0.007* 0.000 0.001 –0.015*** –0.016***
Male –0.320*** –0.355*** 0.072 0.015 0.125* 0.168* 0.126* 0.152*
Black –0.256** –0.099 0.038 0.037 –0.062 –0.024 –0.212* –0.327*
Hispanic –0.210* –0.154 –0.201† –0.209 –0.026 –0.004 –0.106 –0.026
Other race –0.232 –0.356 –0.073 0.069 0.116 0.143 –0.011 –0.030
College –0.163* –0.164* 0.100 0.106 0.112† 0.141† 0.212** 0.180*
Income –0.001 –0.001* 0.000 –0.001 0.000 0.000 –0.001* 0.001†
Married –0.073 –0.045 –0.118 –0.145† 0.056 0.034 –0.061 –0.065
Any children –0.002 0.033 –0.021 –0.016 –0.005 0.009 0.026 0.041
Retired –0.043 –0.053 0.069 –0.004 –0.328*** –0.278** –0.020 0.032
News reliability overall (compared with no change)
Less reliable now –0.025 0.021 0.041 0.117
More reliable now –0.183† –0.125 –0.130 –0.293***
Partisanship
Conservatism 0.010 0.036 0.016 –0.067*
Voted for Trump 0.069 –0.307** –0.046 –0.169†
Voted for someone other
than Trump or Clinton
0.263* –0.118 0.245 –0.023
Did not vote –0.147 –0.048 0.091 0.152
NOTE: Throughout this report, regression models without partisanship have N = 2,649, and with partisanship N = 2,030. † p < 0.10;
* p < 0.05; ** p < 0.01; *** p < 0.001.

standard deviations) for the demographic and other variables used in
this study are presented in Appendix B.
People who reported primarily getting their news from social
media and in-person contacts tended to be younger, female, and white
(versus black or Hispanic), and they tended to have less education
than a college degree. People who reported mostly getting their news
through the print and broadcast television platforms were less likely
to be Hispanic. Radio consumers were significantly more likely to be
male, less likely to be retired, and marginally significantly more likely
to have a college degree. Online news consumers were significantly
younger; more likely to be male, have a college degree, and have lower
income; and less likely to be black.
Political Characteristics and News Consumption Profiles
We know from previous research that political beliefs and partisanship
also affect news consumption decisions. To explore the relationship
between news consumption profiles and political characteristics, we
expanded our second set of analyses to factor in two political variables
taken from the PEPS conducted in 2016. Using the combined data, we
can expand our understanding of media consumption patterns. The
PEPS data feature information on an individual’s political ideology
Table 3.2
Summary of Media Consumption Profile Characteristics: Demographics
Only
Demographics
Social Media/
In-Person
Print/
Broadcast Radio Online
Age Younger N/A N/A Younger
Gender Female N/A Male Male
Race/ethnicity White Non-Hispanic N/A Not black
Educational
attainment
No college
degree N/A College degree College degree
Income N/A N/A N/A Lower income
Married N/A N/A N/A N/A
Retired or
working N/A N/A Working N/A
NOTE: N/A indicates that variables were not significantly associated with news
consumption profiles in the OLS regressions.

(along a 5-point scale, ranging from 1 = “very liberal” to 5 = “very
conservative”) and who they voted for in the 2016 presidential elec-
tion, if they voted (self-reported).6 We were able to link 82 percent of
our 2018 sample to the earlier PEPS sample’s information (N = 2,181)
based on the overlap between the two survey samples within the ALP
overall.7 Table 3.3 presents descriptive results of the analysis, including
the political variables, taken from Model 2 of Table 3.1. Comparisons
are made with people who voted for Clinton because that was the most
commonly reported behavior in the sample (41.2 percent of respon-
6 The correlation between political ideology and voting for Donald Trump or Hillary Clin-
ton was moderate at 0.57 and at –0.056, respectively; however, the results were not substan-
tively different when models were run with each aspect of partisanship separately.
7 Note that restricting analyses to only those who could be matched to the PEPS provides
similar results throughout the analyses to analyses that do not link to the PEPS.
Table 3.3
Summary of Media Consumption Profile Characteristics (Considering
Partisanship)
Demographics
Social Media/
In-Person Print/Broadcast Radio Online
Age Younger Older N/A Younger
Race or ethnicity N/A N/A N/A Not black
Educational
attainment
No college
degree
N/A College degree College degree
Income Lower income N/A N/A Higher income
Married N/A Not married N/A N/A
Retired or working N/A N/A Working N/A
Partisanship
Political
ideology
N/A N/A N/A More liberal
Vote choice in
2016
Most likely to
have voted for
someone other
than Clinton or
Trump
Most likely to
have voted for
Clinton
N/A Most likely to
have voted for
Clinton
NOTE: N/A indicates that variables were not significantly associated with news
consumption profile factors in the OLS regressions.

dents, followed by 35.1 percent who voted for Trump) and because
having the largest group as the referent category simplifies the analysis
and interpretation. When reporting these results, however, we consider
comparisons of different groups of voters (those who voted for Clinton,
those who voted for Trump, those who voted for someone else, and
those who did not vote). We do not report every comparison in every
case; we emphasize those comparisons that are statistically significant
and meaningful while pointing out key areas where no clear relation-
ship exists.
We find that political beliefs and voting behavior matter, but not
as much as might be expected based on previous work and popular
news coverage. First, compared with those who voted for Clinton in
2016, respondents who reported voting for Trump were significantly
less likely to score as highly on the print/broadcast news consump-
tion profile. However, Trump voters were no more or less likely to
score differently on any of the other news consumption profiles.
Second, those who voted for someone other than Trump or Clinton
were more likely than those who voted for Clinton to score highly
on the social media/in-person profile. Third, respondents who rated
their political ideology as more conservative were somewhat less likely
to be in the online group (and those who self-identified as liberals
slightly more likely), although this result is both substantively smaller
and marginally statistically significant. The addition of the two politi-
cal variables generally does not substantially alter the previous find-
ings related to the demographics of the news consumption profiles.
Exceptions are that when political characteristics are also considered,
those who were married were less likely to score highly in the print/
broadcast consumption profile, older individuals were more likely to
score highly in the print/broadcast profile, and low income was associ-
ated with higher scores on the social media/in-person profile.
Taken together, older individuals were more likely to report get-
ting their news using more-traditional platforms—print and broad-
cast television—than younger individuals. Younger respondents were
more likely to report getting their news from social media/in-person
sources or online. Women were more likely to report getting their news
through more-social means (social media or in person). Non-Hispanic

white respondents were generally more likely than others to report rely-
ing heavily on social media/in-person sources and online platforms for
obtaining their news.
Attitudes About News Reliability
Perceptions of Whether News Has Become More or Less Reliable
We next investigate whether there are links between the ways in which
people access news and their perceptions of news reliability. That is,
are there consistent differences in perceptions of the reliability of news
across media consumption profiles? In addition to reporting which
sources they use most often, respondents were asked to rank “which
source provides the most-reliable information, in your opinion” and
whether they “find the news you currently receive . . .” to be “more
or less reliable . . . than in the past” (or about the same). We asked
about reliability of information, rather than trust, because we wanted
to assess the degree to which respondents felt that they could rely or
consistently depend on a given news platform and its coverage to pro-
vide accurate information. The term reliability captures this consistent
dependability more directly than the word trust and places the empha-
sis on news provided by a media source, which was what we were most
interested in, rather than on the institution itself. We note that past
research on perceived media credibility or trust has tended to use the
terms interchangeably but has also demonstrated that the various com-
ponents of media trust (such as accuracy, fairness, unbiasedness, trust-
worthiness, reliability) load together well and are empirically insepa-
rable.8 We explore the relationship between the concepts of reliability
and trust further in a subsequent section, and we show that results are
qualitatively the same using “reliability” or “trust.” To do so, we added
8 Richard Fletcher and Sora Park “The Impact of Trust in the News Media on Online
News Consumption and Participation,” Digital Journalism, Vol. 5, No. 10, 2017; Gaziano,
and McGrath, 1986; Philip Meyer, “Defining and Measuring Credibility of Newspapers:
Developing an Index,” Journalism Quarterly, Vol. 65, No. 3, 1988; Tsfati and Cappella,
2003; Mark Douglas West, “Validating a Scale for the Measurement of Credibility: A Cova-
riance Structure Modeling Approach,” Journalism Quarterly, Vol. 71, No. 1, 1994.

consideration of level of trust toward a variety of news platforms from
a separate ALP survey. Further details of the comparison are presented
in Appendix C, but the results show that, conceptually, trust and reli-
ability are closely related for consumers when assessing various news
platforms. Furthermore, just as perceived reliability is associated with
news consumption choices, a respondent’s trust in various news plat-
forms is also associated with information use habits.
General perceptions about the reliability of the news overall
among the survey respondents reflected a substantial amount of pes-
simism. Although 44 percent reported that they believed “the news is
as reliable now as in the past,” nearly the same amount—41 percent—
reported a belief that the news has become less reliable. A minority
(15 percent) said that they believed that the news is more reliable now
than in the past. There are systematic differences in perceptions based
on both demographic and political characteristics (Tables 3.4 and 3.5).
Table 3.4 presents the relative risk ratios derived from a multinomial
logit regression, comparing those reporting that news is “more reli-
able now” and those reporting that the news is “less reliable now” to
those reporting that the news is “as reliable now.” Values greater than 1
indicate factors associated with increased likelihood of the belief (com-
pared with “as reliable now”); values less than 1 identify characteristics
that are associated with a reduced likelihood of the belief. Table 3.5
presents a verbal summary of the results in Table 3.4.
Without attention to partisanship (Model 1 in Table 3.4), respon-
dents who were white, male, or retired or who had higher incomes or
less than a college education were significantly more likely to believe
the news is less reliable now (compared with as reliable as in the past).
Conversely, women, racial or ethnic minorities, and those without col-
lege degrees were significantly more likely to say they believed that
the news is more reliable now than in the past. Finally, we sought to
explore the relationship between news consumption profiles and per-
ceptions of reliability. We found that those who were more likely to use
social media/in-person sources and online news platforms to get their
news were less likely to feel that news had become more reliable, but
no profile was associated with perceptions of reductions in reliability.

Table 3.4
Characteristics Associated with Perceptions of News Reliability (Relative
Risk Ratios from Multinomial Logistic Regression)
Demographics
More Reliable Now Less Reliable Now As Reliable Now
Model 1 Model 2 Model 1 Model 2 (Reference)
Age 1.003 0.996 0.992 0.988
Male 0.632* 0.799 1.446* 1.323
Black 1.959* 2.277* 0.393** 0.824
Hispanic 2.204** 2.419** 0.634† 0.872
Other race 1.078 0.695 0.699 0.740
College 0.613† 0.602* 0.658* 0.949
Income 1.004 1.001 1.006*** 1.003†
Married 0.942 0.694 0.901 0.906
Any children 0.776 0.761 0.991 0.802
Retired 0. 839 0.799 1.551* 1.462†
News consumption profile
Social media/
in-person
0.840† 0.806* 1.030 0.981
Print/
broadcast
0.950 0.884 1.041 1.025
Radio 0.947 0.855† 1.069 1.040
Online 0.747*** 0.703*** 0.985 1.139
Partisanship
Conservatism 0.942 1.112
Voted for
Trump
0.840 3.508***
Voted for
someone
other than
Trump or
Clinton
1.003 3.373***
Did not vote 0.127*** 0.700
† p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001.

Although political variables were a relatively minor factor in our
analysis of news consumption profiles, they played a significant role in
our assessment of perceptions of news reliability. Specifically, when we
factored in the political variables (Model 2 in Table 3.4), we found that
people who did not vote were less likely than others to report believ-
ing that the news is more reliable now than in the past; compared
with Clinton voters, those who voted for anyone else were more than
Table 3.5
Summary of Associations Between Consumption Profiles and Perceptions
of News Reliability
Demographics Perceptions
Age N/A
Gender Men are less likely to think the news is more reliable
now and more likely to think it is less reliable.
Race/ethnicity Black and Hispanic respondents are more likely than
white respondents to think the news is more reliable
now and less likely to think it is less reliable.
Educational attainment Those with college degrees are less likely than those
without to think the news is more reliable now.
Income Those with higher incomes are more likely to think
the news is less reliable now.
Married N/A
Retired or working Retired people are more likely than working people
to think the news is less reliable now.
Partisanship
Political ideology N/A
Voted in 2016 People who voted for someone other than Clinton
are more likely to think the news is less reliable now,
and people who voted for anyone (compared with
nonvoters) think the news is more reliable now.
Social media/in-person People who rely less on social media/in-person think
the news is more reliable now.
Print/broadcast N/A
Radio People who rely less on radio think the news is more
reliable now.
Online People who rely less on online news think the news
is more reliable now.
NOTE: N/A indicates that variables were not significantly associated with
perceptions of news reliability in the multinomial logistic regression.

three times as likely to report a perception that the news is less reliable
than in the past. We found no relationship between perceived reliabil-
ity and political ideology when voting behavior was also considered.9
The inclusion of the political variables also affects which sociodemo-
graphic characteristics are statistically significant. Nearly all of the
demographic factors predicting perceptions of reduced reliability, apart
from income and retired status, are absorbed by the political variables,
confirming the strong relationship between partisanship and demo-
graphic characteristics.10 News consumption profiles remain relevant,
however, with the inclusion of the partisanship variables. Essentially,
the more heavily that individuals relied on social media/in-person,
radio, or online platforms, the less likely they were to believe that the
reliability of the news had changed for the better, although there were
no profile links to perceptions of deteriorating reliability.
Which Platforms Are Considered Most Reliable
To further explore media reliability, we asked respondents to report
which types of media platforms they found most reliable. Unsurpris-
ingly, perceived reliability differed. Based on our survey, broadcast and
cable television were perceived by the most people to be the most reli-
able, with most people rating one or both in the top two (Figure 3.3).
Television is followed by print, online news sites, and radio. Social
media and in-person news sources were perceived as the most-reliable
sources by the smallest number of respondents.
There are systematic demographic differences in who rated which
sources as the most reliable. Comparisons are made with broadcast
television because it was most commonly reported as the most reliable
platform (26.3 percent of respondents). Tables 3.6 and 3.7 summa-
rize the results: Table 3.6 presents the results of a multinomial logis-
tic model considering the likelihood of rating a platform other than
9 When political ideology was considered without also controlling for the 2016 presidential
vote, more-conservative individuals were more likely to report “less reliable.”
10 For a treatment of the relationship between partisanship and demographics, see Liliana
Mason, Uncivil Agreement: How Politics Became Our Identity, Chicago, Ill.: University of
Chicago Press, 2018.

broadcast television as the most reliable platform using demographic,
political, and overall perceptions of news reliability; Table 3.7 provides
a verbal summary of the relationships.
Demographic and Political Characteristics
Those who viewed print as the most reliable news platform (compared
with broadcast television) were significantly more likely to have college
degrees and higher incomes, and to be retired and male. Race also mat-
ters: Respondents who were black were less likely to list print as the most
reliable source of media information. People who voted for candidates
other than Trump in 2016 were more likely than those who did vote for
Trump to report that print media is the most reliable news platform.
Cable television is significantly distinguished from broadcast tele-
vision as the more reliable television news platform by those who voted
for Trump; compared with those who voted for anyone else, Trump
voters were more than twice as likely to report that cable news is more
reliable than broadcast news. Conversely, those who voted for Clin-
ton or third-party candidates were less likely to report cable news as
the most reliable source of news and might value other sources more
Figure 3.3
Respondents’ Top Two Perceived Most-Reliable News Sources
Percentagelistingsource
asfirstorsecondchoice
0
5
10
15
20
25
30
35
40
45
50
media
In person
Source of news
34
47
40
23
32
13 13

Table 3.6
Perceived Most Reliable News Platform, Compared with Broadcast Television (Relative Risk Ratios)
Characteristic Print Cable Television Radio Online Social Media In-Person
Age 0.990 1.019 0.992 0.969** 1.008 0.992
Male 1.579* 1.438 2.712*** 1.944** 0.775 0.933
Black 0.386* 0.988 0.507 0.261* 1.655 1.165
Hispanic 1.199 1.025 0.790 0.589 5.691*** 0.619
Other race 1.338 2.007 3.263* 1.392 1.060 0.094*
College 2.224*** 0.945 1.161 2.178** 0.541 1.779
Income 1.007*** 1.001 1.004 1.000 0.994 0.994
Married 0.706 1.000 0.680 0.965 0.975 2.800*
Any children 0.681 0.963 1.130 1.299 0.855 0.838
Retired 1.700† 1.148 0.569† 1.322 0.313† 0.948
Less reliable now 1.680* 0.961 1.490 1.571 4.253* 4.713***
More reliable now 0.867 2.248** 0.716 1.653 3.342† 2.151
Partisanship
Conservatism 0.953 0.909 0.933 0.842 1.043 1.075
Voted for Trump 0.296*** 2.256** 1.454 1.017 3.337* 1.404
1.555 0.940 1.777 4.003*** 0.701 5.623**
Did not vote 0.462 1.488 0.549 1.686 4.368* 2.638
† p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001.

Table 3.7
Summary of Associations Between Consumption Profiles and Most Reliable
Platform: Demographic and Political Characteristics
Demographics Characteristic
Age Older people are more likely to rate online news as less reliable
than broadcast television.
Gender Men are more likely to rate radio, online, and print news as
more reliable than broadcast television.
Race or ethnicity Black respondents are less likely than white respondents to rate
print and online as more reliable than broadcast television;
Hispanic respondents are more likely than others to rate social
media as the most reliable platform; and those who are not
white, Hispanic, or black are more likely to rate radio as the
most reliable and least likely to rate in-person as most reliable.
Educational
attainment
Those with college degrees are more likely than others to rate
print or online news as more reliable than broadcast television.
Income Those with higher incomes are more likely to rank print news
as most reliable.
Married Married people are more likely than others to rate in-person
news as the most reliable.
Retired or working Retired people are more likely than others to rate print as most
reliable, and less likely to rate radio or social media as most
reliable.
Partisanship
Political
ideology
N/A
Voted in 2016 Compared with Clinton voters, Trump voters are more likely
to rate cable television and social media as more reliable, and
print as less reliable. People who voted for someone other than
Trump or Clinton are more likely to rate online and in-person as
most reliable, and nonvoters rate social media as most reliable.
Less reliable
now
People who think the news has become less reliable are more
likely to rate print, social media, and in-person as most reliable.
More reliable
now
People who think the news has become more reliable are more
likely to rate cable television and social media as most reliable.
NOTE: N/A indicates that variables were not significantly associated with ratings of
most reliable news platform in the multinomial logistic regression.

highly. No other political or demographic characteristics significantly
distinguished those who reported cable television as the most reliable
form of news from other respondents.
Radio was more likely to be reported as the most reliable platform
by men and by those who identified as something other than white,
black, or Hispanic. Retired individuals were less likely than those who
were not retired to identify radio as the most reliable platform.
Younger people, men, those who were white (relative to black),
and those with college degrees were more likely to report online plat-
forms to be the most reliable (compared with broadcast television).
In contrast, Hispanic individuals were substantially more likely than
others to list social media as the most reliable news platform. Respon-
dents who voted for anyone other than Trump in 2016 were signifi-
cantly less likely than Trump voters to list social media as the most
reliable news platform.
Finally, obtaining news in person was rated as the most reliable
source significantly more by married individuals (nearly three times
more than unmarried individuals), and those who voted for someone
other than Clinton or Trump (more than five times as likely as Clin-
ton voters). Individuals who identified as something other than white,
black, or Hispanic were much less likely than others to list in-person as
the most reliable platform.
Perceptions of Overall News Reliability
We also observe links between perceptions of the overall reliability
of the news and ratings of the most-reliable news platforms. People
who reported believing that the news is less reliable now were more
likely to rate print, social media, or in-person news as the most reliable
(compared with broadcast television); those who reported believing the
news to be more reliable now than in the past were more likely to rate
cable television or social media as the most-reliable platforms (com-
pared with broadcast television).
Although demographic and political characteristics were important
in predicting perceived reliability of specific platforms, consumption
profiles were also significant, suggesting a clear relationship between

the perceived reliability and decisions about consumption (though we
cannot say from our analyses which way the arrow moves, from per-
ceived reliability to consumption choices or from consumption choice
to perceived reliability. It is likely a mix of both). Table 3.8 presents
the expanded multinomial logistic regression results, and Table 3.9
verbally summarizes our analysis. In general, the results suggest that
people tend to rate most highly those sources that they use, consistent
with prior studies. However, there are some other interesting relation-
ships. For example, those in the radio and online consumption profiles
tended to rate almost every other news platform higher than broadcast
television news (despite it being the most trusted overall); those in the
print and broadcast category were least likely to rate cable television,
social media, and online sources of news as the most reliable. We con-
sider these findings in more detail.
Perceived Reliability’s Influence on News Consumption Choices
To further explore the relationship between perceived reliability and
media use, we compared each respondent’s most-used sources with
those they rated as the two most reliable. Our analysis indicates that
most people (72 percent) get news from sources they believe are among
the most reliable. That is, one or both of the platforms they use most
frequently are also perceived to be among their top two most-reliable
platforms. However, that still leaves 28 percent, almost one-third, of
respondents who get most of their news from platforms they have not
rated as the most reliable.
Table 3.10 presents results from logistic regressions predicting that
individuals use platforms they rate as most reliable, and Table 3.11 pres-
ents a verbal summary. There are few demographic and no political char-
acteristics strongly associated with consumption of self-identified reliable
sources. There is some evidence that race might matter in whether indi-
viduals consume news from sources that they identify as most reliable,
but the two models tell slightly different stories about how.11 Notably, if
we do not consider political characteristics, marriage is negatively associ-
ated with getting news from a reliable source. In other words, married
11 Again, restricting to only those with partisanship information produces similar results.

Table 3.8
Perceived Most Reliable News Platform, Compared with Broadcast Television, Including News Consumption Profile
(Relative Risk Ratios)
Characteristic Print
Cable
Television Radio Online Social Media In-Person
Age 0.994 1.015 1.001 0.992 1.040† 1.012
Male 1.349 1.112 2.453** 1.682† 0.963 1.063
Black 0.448 1.086 0.578 0.337 2.321 2.000
Hispanic 1.240 0.807 0.688 0.560 7.121*** 0.932
Other race 1.242 1.590 3.348* 1.512 1.697 0.109†
College 1.819* 0.837 0.935 1.810† 0.554 1.640
Income 1.007*** 1.000 1.004 0.999 0.994 0.997
Married 0.716 0.881 0.736 1.018 1.030 2.792*
Any children 0.651 0.965 1.114 1.154 0.899 0.790
Retired 1.926* 1.198 0.707 1.424 0.330† 1.349
Less reliable now 1.613† 0.966 1.438 1.454 4.132* 5.135***
More reliable now 0.978 1.982* 1.073 2.660* 4.349* 2.234

NewsThatPeopleConsume:NewsConsumptionProfiles45
Table 3.8—Continued
Characteristic Print
Cable
Television Radio Online Social Media In-Person
Social media/in-person 0.869 0.569*** 0.823 1.039 2.151* 2.583**
Print/broadcast 1.206 0.433*** 0.955 0.725* 0.709† 0.743
Radio 2.019*** 1.516** 5.511*** 2.084*** 0.988 5.611***
Online 1.676*** 1.092 1.787*** 3.452*** 1.778*** 0.793
Partisanship
Conservatism 0.902 0.894 0.934 0.862 1.069 1.044
Voted for Trump 0.381** 2.402** 1.950 1.246 3.752* 1.082
Voted for someone other than Trump
or Clinton
1.609 1.049 1.363 4.373*** 0.650 4.491**
Did not vote 0.379 1.239 0.278 1.365 5.373* 1.442
† p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001.

people are less likely to identify as most reliable those sources that they
report using most heavily.
There is also evidence that perceptions of the overall reliability
of the news is linked to whether people primarily get their news from
reliable platforms; individuals who believe that the news is less reliable
now than in the past are less likely to get their news via platforms they
deem most reliable than either those who perceive no change in the
overall reliability of the news or those who believe the news is more
reliable now.
The news consumption profile variables are more definitive.
Those in the social media/in-person group are less likely to get their
news from sources they identify as most reliable (in other words, they
rely on social media and other people for news but do not consider
these to be the most reliable); those in the print/broadcast profile are
more likely to rely on sources they identify as reliable.
Our analysis suggests that although perceived reliability is usu-
ally associated with decisions about news consumption, the two are
not always related. For example, individuals who believe that overall
the news is less reliable now than in the past are about one-third less
likely than others to primarily obtain news from platforms they rate
Table 3.9
Summary of Associations Between Consumption Profiles and Perceived
Most Reliable Platform
Consumption Profile Characteristic
Social media/
in-person
People in this profile are more likely to rate social media and
in-person as most reliable and cable television as less reliable.
Print/broadcast People in this profile are more likely to rate cable, online, and
social media as less reliable.
Radio People in this profile are more likely to rate broadcast
television and social media as less reliable than print, radio,
online, and in-person
Online People in this profile are more likely to rate broadcast
television, cable television, and in-person as less reliable than
print, radio, online, and social media.
NOTE: N/A indicates that variables were not significantly associated with rating of
most reliable news platform in the multinomial logistic regression.

Table 3.10
Primarily Gets News from One of Their Top Two Perceived Most-Reliable
Platforms (Odds Ratios; vs. Does Not)
Characteristic Model 1 Model 2
Age 1.003 1.006
Male 1.268 1.192
Black 0.601† 0.825
Hispanic 0.800 0.720
Other race 0.681 0.428*
College 0.798 0.739
Income 1.000 0.999
Married 0.681* 0.738
Any children 0.826 1.001
Retired 1.158 1.350
Less reliable now 0.735† 0.642*
More reliable now 1.086 1.133
Social media/in-person 0.693*** 0.753**
Print/broadcast 1.159* 1.141†
Radio 1.022 1.075
Online 0.881 0.947
Partisanship
Conservatism 0.938
Voted for Trump 1.313
Voted for someone other than Trump or Clinton 1.691
Did not vote 0.936
† p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001.

Table 3.11
Summary of Variables Linked to an Individual’s Likelihood of Getting News
from at Least One of Their Two Perceived Most-Reliable Platforms
Age N/A
Gender N/A
Race/ethnicity Those who are not white, black, or Hispanic are less likely to
get news from platforms they rate as most reliable.
Educational
attainment
N/A
Income N/A
Married N/A
Retired or working N/A
Partisanship
Political ideology N/A
Voted in 2016 N/A
Less reliable now Those who report that news is less reliable today than in the
past are less likely to get news from platforms they rate as
most reliable.
More reliable now N/A
Social media/
in-person
People who rely on social media/in-person more are less
likely to get news from platforms they rate as most reliable.
Print/broadcast People who rely on print/broadcast more are more likely to
get news from platforms they rate as most reliable.
Radio N/A
Online N/A
NOTE: N/A indicates that variables were not significantly associated with the
likelihood of primarily getting news from self-rated reliable platforms in the logistic
regression.

as the most reliable. Rather than rely more heavily on news platforms
they believe to be the most reliable as a means to mitigate perceived
decreases in overall news reliability, these individuals are at greater risk
of doing the contrary. Furthermore, as noted, social media/in-person
news consumers are less likely to rely primarily on those sources that
they rate as most reliable. They do not necessarily view social media or
in-person sources as among the most reliable and yet still turn to these
sources most often to get news. This suggests that—at least for this
group of individuals (typically younger, female, white, and without a
college degree)—news consumption might be driven less by perceived
reliability of information and more by other factors. We can only spec-
ulate on these factors, but literature suggests they might include inter-
est, time, or willingness. As an example, we found that women were
more likely to rely on social media and in-person channels but were not
more likely to rate these channels as most reliable. One possible expla-
nation (based on previous research on women, their daily demands,
and their personal networks) is that women might find it more conve-
nient to be informed about news through social channels (in-person,
social media) that are more suited to their larger personal networks,12
higher levels of communication across those networks,13 and relative
lack of leisure time compared with men.14 Although this is consistent
with the empirical results, it would need to be explored in more depth
to directly support any conclusions.
Prior studies that similarly identify a disconnect between cred-
ibility and frequency of use among some individuals provide a vari-
ety of additional explanations for why people might rely on sources
12 Steve McDonald and Christine A. Mair, “Social Capital Across the Life Course: Age and
Gendered Patterns of Network Resources,” Sociological Forum, Vol. 25, No. 2, June 2010.
13 Carolyn A. Liebler and Gary D. Sandefur, “Gender Differences in the Exchange of Social
Support with Friends, Neighbors, and Co-Workers at Midlife,” Social Science Research, Vol. 31,
No. 3, 2002.
14 Max Haller, Markus Hadler, and Gerd Kaup, “Leisure Time in Modern Societies: A New
Source of Boredom and Stress?” Social Indicators Research, Vol. 111, No. 2, 2013.

other than those they identify as the most reliable.15 Tsfati and Cap-
pella argue that people might get news from sources they believe to
be unreliable because, in addition to gathering factual information,
news consumption might serve other needs, such as social needs (to
stay in touch with other people or society at large) or a need for inter-
esting entertainment.16 They also argue that the psychological “need
for cognition”—the need to think, to understand, to make sense of the
world, and to learn about various points of view—motivate consump-
tion regardless of whether the source is perceived as trustworthy.17 This
is another area where additional research would be valuable.
Who Seeks Out Differing Views and How Often
An additional aspect of news consumption behavior we examined was
how often individuals sought out sources of news that they knew would
offer views that were different from their own. Our inclusion of this
question was motivated by recent attention to the negative effects of
echo chambers and filter bubbles18—and given this attention, it should
be noted that this is one case in which respondent answers might be
heavily affected by desirability bias because respondents might not
want to admit that they do not look for alternative views.19 Thus, it
is possible that respondents will overreport the frequency with which
they look for alternative viewpoints; this is an important caveat to keep
in mind when assessing these results.
In our survey, one in five respondents (20 percent) reported that
they “always or almost always” sought out different views; an addi-
tional 54 percent said that they did so “sometimes.” Only 17 percent
said that they “infrequently” sought out differing sources, and 9 per-
cent said “never or almost never.”
15 Fletcher and Park, 2017; Tony Rimmer and David Weaver, “Different Questions, Differ-
ent Answers? Media Use and Media Credibility,” Journalism Quarterly, Vol. 64, No. 1, 1987.
16 Tsfati and Cappella, 2003.
17 Tsfati and Cappella, 2003.
18 For example, see Garrett, 2009.
19 Paul Lavrakas, “Response Bias,” Encyclopedia of Survey Research Methods, Thousand
Oaks, Calif.: Sage, 2008.

Table 3.12 presents the results of a multinomial logistic regres-
sion analysis linking how often people reported seeking out news
from viewpoints that differ from their own, compared with “some-
times” or “infrequently.” A verbal summary of the results is presented
in Table 3.13. Married people are particularly less likely to report that
they “always or almost always” sought out opposing news sources;20
black and other race respondents were significantly more likely than
others to report “never or almost never.” Those with higher education
and higher incomes were significantly less likely than others to report
they “never or almost never” sought out differing views.
Perceptions of the overall reliability of the news are also linked to
whether people seek out alternative viewpoints in the news they con-
sume. People who believe the news to be more reliable now than in the
past are more likely to report that they “always or almost always” seek
out alternative views, and people who report any change in the overall
reliability of the news (either an increase or a decrease in reliability) are
less likely to report that they “never or almost never” seek out alterna-
tive views than those who think the news is as reliable now as in the
past.
Political ideology and vote choice also matter. In terms of seeking
out differing views, people who rated their political ideology as more
conservative were less likely than more-liberal respondents to report
that they “never or almost never” did so; people who reported voting
for someone other than Clinton were roughly three times as likely as
Clinton voters to report that they “always or almost always” seek out
differing views in their news. Those who voted for Trump were three
times as likely as non-Trump voters to report “never or almost never.”
This suggests that Trump voters place themselves in opposite catego-
ries of people who either always or never seek out differing views; Clin-
ton voters were much more likely to “sometimes” or “infrequently”
seek out differing views than others.
20 Note that this is in opposition to the explanation in Tsfati and Cappella, 2005, of the
need for understanding alternative viewpoints as one potential reason for the consumption
of news from unreliable sources; married people are less likely to get news from self-identified
reliable sources but are also less likely to seek out alternative viewpoints.

Table 3.12
How Often Respondent Seeks Out News with Different Views (“Always
or Almost Always” or “Never or Almost Never” vs. “Sometimes” or
“Infrequently”; Relative Risk Ratios)
Characteristic Always or Almost Always Never or Almost Never
Age 0.996 1.005
Male 1.849** 1.206
Black 1.957* 2.958*
Hispanic 0.998 1.451
Other race 1.050 2.872†
College 1.215 0.446*
Income 0.999 0.995†
Married 0.534** 0.899
Any children 1.095 0.947
Retired 0.955 1.023
Less reliable now 1.434 0.271**
More reliable now 1.472† 0.416**
Social media/in-person 0.903 1.128
Print/broadcast 1.139 0.844
Radio 1.147 1.022
Online 1.440** 0.732**
Partisanship
Conservatism 0.946 0.784†
Voted for Trump 2.890*** 3.617**
3.407*** 0.575
Did not vote 1.643 1.160
† p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001.

Table 3.13
Summary of Characteristics Associated with Likelihood of Seeking Out
News with Different Views
Age N/A
Gender Men are more likely to “always or almost always” seek out differing
views.
Race or
ethnicity
Black respondents are more likely than white respondents to
“always or almost always” or “never or almost never” seek out
differing views. Those who are not white, black, or Hispanic are
more likely than white respondents to “never or almost never” seek
out differing views.
Educational
attainment
Those without college degrees are more likely to “never or almost
never” seek out differing views.
Income Those with lower incomes are more likely to “never or almost never”
seek out differing views.
Married Married people are less likely than others to “always or almost
always” seek out differing views.
Retired or
working
N/A
Partisanship
Political
ideology
More-liberal people are more likely to “never or almost never” seek
out differing views.
Voted in
2016
People who voted for Trump are more likely than Clinton voters
to “always or almost always” or “never or almost never” seek out
differing views. Those who voted for someone other than Trump
or Clinton are more likely than Clinton voters to “always or almost
always” seek out differing views.
Less reliable
now
People who think the news is now less reliable are less likely to
“never or almost never” seek out differing views.
More
reliable now
People who think the news is now more reliable are more likely to
“always or almost always” and less likely to “never or almost never”
seek out differing views.
Social media/
in-person
N/A
Print/
broadcast
N/A
Radio N/A
Online People who rely on online news are more likely to “always or almost
always” and less likely to “never or almost never” seek out differing
views.
NOTE: N/A indicates that variables were not significantly associated with the likelihood
of seeking out news with different viewpoints in the multinomial logistic regression.

Turning to the relationship between likelihood of seeking out
alternative viewpoints and news consumption profiles, we found that
people who scored highly in the online news consumption profile were
significantly more likely to report “always or almost always” seeking
out sources of news that offer views different from their own than was
the case for other consumer profiles. Online consumers were also sig-
nificantly less likely to report that they “never or almost never” sought
out differing views. It is possible this reflects the ease with which
information consumers can seek out diverse viewpoints online, if they
choose to do so.
Trust in Institutions and Media Consumption
Limitations
Our analyses have a few limitations worth mentioning. First, we only
asked in our survey about news consumption at the level of news plat-
form, not at the level of individual outlets. The quality of news an
individual gets through any platform can vary substantially depending
on the specific outlets consumed. For example, a consumer who relies
on the internet platform for news but is using it to access the websites
of leading newspapers will likely receive a very different information
diet from someone who uses the internet to access highly polarized
editorials or blogs. We hope to expand our work to the outlet level in
the future.
Second, we know little about specific frequency of use. We did
not ask respondents how frequently or for what duration they typi-
cally use the different news platforms, only the rank ordering of use
frequency. As noted, a consumer who reads hours of news from news-
papers every day will have a very different information diet than a user
who reads the newspaper once a month, even if that is still the primary
news source for that consumer. Incorporating time use at the outlet
level into future iterations of this survey would be valuable.
Third, our analysis applies only to English speakers, as the quar-
terly ALP Omnibus survey we used to field our questions was not
translated into Spanish for Spanish-only respondents. Despite this

limitation, our sample was 14 percent Hispanic (unweighted). Future
work should seek to explore news consumption habits among Spanish-
speaking (and other non-English-speaking) populations in the United
States.
Fourth, we interpret the results in terms of non-entertainment
news. The survey items referred to “national and international news”
and examples of outlets within each platform identified primarily non-
entertainment news outlets (e.g., Time or Newsweek magazine, CNN,
Breitbart), but we did not explicitly tell respondents to exclude enter-
tainment information consumption. It is possible some respondents
included entertainment news in their responses. Consumption behav-
iors and platform choices related to entertainment versus hard news
might differ, and we are unable to further distinguish results.
Finally, our results are dependent on the ways in which our
respondents interpret and respond to questions. Social desirability bias,
described previously, could affect responses about seeking out other
sources of information and about perceived reliability of specific plat-
forms. Individuals might overreport consumption of what they believe
to be higher-prestige media (print versus social media, for example).
Furthermore, individuals might have an interest in identifying as most
reliable those sources that they tend to rely on most often, to main-
tain internal consistency, even if this response is not entirely truthful.
Respondents likely also do not spend a lot of time thinking on a daily
basis about how they will assess their degree of trust in media outlets
or even ranking their use of different platforms. Their responses on
the survey might reflect reactions to the questions without necessarily
being indicative of longer-term beliefs and assessments. Overcoming
this type of bias is challenging, but in future iterations we might con-
sider using open-ended response questions to get a better sense of why
individuals hold the beliefs they do or asking similar related questions
at the beginning and end of a longer survey to assess consistency and
durability of responses.

57
CHAPTER FOUR
Discussion
Four Motivating Questions
This study of U.S. news consumers’ relationship with the changing
U.S. news environment—their preferred news outlets, their sense of
the reliability of those and other news outlets, their willingness to seek
out news from differing viewpoints, and their demographic and politi-
cal partisan characteristics connecting them to these aspects of news
consumption—was motivated by four interrelated questions:
• How do Americans get their news?
• How are news consumption choices linked to demographic or
• Do news consumers believe the reliability of news has changed,
and which news platforms do they believe to be more or less reli-
able?
• How is the perceived reliability of news associated with news con-
sumption choices?
Overall Findings
Overall, we found that broadcast television, followed by online sources,
cable television, and social media are the most-common platforms on
which news is consumed by English-speaking U.S. adults.

From our sample of respondents, we identified four common news
consumption profiles (described and differentiated by people’s relative
level of use of four different packages to obtain news): print and broad-
cast television platforms, online platforms, radio, and social media and
in-person sources. An individual’s use of various sources of news is rep-
resented by how highly he or she scores on each of these four profiles.
These news consumption profiles are useful for several reasons.
Most importantly, they provide a more holistic picture of news con-
sumers, allowing us to identify which types of sources are often used
together and the specific demographic characteristics associated with
specific patterns of media consumption. They also allow us to explore
how news consumption fits in with other political and social behaviors,
offering a more nuanced view of how news consumption might fit into
individual lives. For example, working individuals tend to score higher
on radio sources (possibly while commuting to work). Married individ-
uals are less likely to rely on print or broadcast news, more likely to per-
ceive in-person communication as the most reliable method of obtain-
ing news, and less likely to report seeking out news that offers different
viewpoints (possibly because of time constraints). As noted elsewhere,
however, we will also need information on specific sources used and
duration of use to fully assess information consumption habits. Col-
lecting and analyzing this information will be an important next step
in this line of research.
Perceived Reliability of News Platforms
Two variables, age and political attitudes and behavior, appear to be
the most-salient predictors of news consumption decisions and per-
ceived reliability of information sources. With respect to age, as dis-
cussed in earlier chapters, younger users are more likely to lean on
online and in-person information channels; older users are more likely
to get news through more-traditional means—broadcast or cable news
and print media. Across users of all ages, however, social media and
in-person communication sources are least likely to be perceived as
reliable sources of news, even by those who rely primarily on them.
Broadcast and cable television were viewed as the most reliable, fol-

Discussion 59
lowed by print and online sources. Tables 4.1 through 4.3 summarize
these and the other overall findings we outline later; Table 4.1 presents
results in terms of characteristics and behaviors associated with each
news consumption profile factor; Table 4.2 summarizes results related
to perceptions of overall reliability of the news; and Table 4.3 summa-
rizes results related to our political measures.
Partisanship and voting behavior are relevant to news consump-
tion choices and perceptions of reliability. People who reported holding
a more conservative political ideology were less likely to score highly
Table 4.1
Summary of Relationships Between News Consumption Profiles and Other
Characteristics and Behaviors
Characteristic
Social Media/
In-Person
Print/
Age Younger N/A N/A Younger
Race or ethnicity White Non-Hispanic N/A Not black
Educational
attainment
No college
degree
N/A College degree College degree
Income Lower income N/A N/A Higher income
Married N/A Not married N/A N/A
Retired or working N/A N/A Working N/A
Partisanship
Political
ideology
N/A N/A N/A More liberal
Vote choice in
2016
More likely to
have voted for
someone other
than Clinton or
Trump
More likely to
have voted for
Clinton
N/A More likely to
have voted for
Clinton
Perception of
change in news
reliability
Does not think
news is more
reliable now
N/A N/A Does not think
news is more
reliable now
Consumption behaviors
Uses reliable
platforms
No Yes N/A N/A
Seeks out
differing views
N/A N/A N/A Yes
NOTE: N/A indicates that variables were not significantly associated with the news
consumption profile factors in the regressions.

on the online consumption and less likely to report “never or almost
never” seeking out alternative views in their news. More-liberal indi-
viduals, conversely, tended to score higher on the online consumption
profile and were more likely to report “never or almost never” seeking
out alternative views than more-conservative respondents.
Compared with people who reported voting for Clinton in 2016,
people who reported voting for Trump scored lower on the print/broad-
cast television profile. Trump voters were also more likely to believe
that the news is less reliable now than in the past and to rate cable tele-
Table 4.2
Summary of Relationships Between Perceptions of Overall News Reliability
and Other Characteristics and Behaviors
Characteristics of People Who
Think the News Is . . . . . . More Reliable Now . . . Less Reliable Now
Race or ethnicity Black and Hispanic N/A
Educational attainment Less than a college degree N/A
Income N/A Higher income
Retired or working N/A Retired
Partisanship
Political ideology N/A N/A
Voted in 2016 Voted (vs. did not vote) Voted for someone other
than Clinton
Social media/in-person Use social media/in-person
less
N/A
Print/broadcast N/A N/A
Radio Use radio less N/A
Online Use online news less N/A
Consumption behaviors
Uses reliable platforms N/A Less likely to use self-rated
reliable platforms
Seeks out differing views Always or almost always Less likely than people
who do not think the
reliability of the news has
changed to say never or
almost never
NOTE: N/A indicates that variables were not significantly associated with
perceptions of the reliability of the news in the regressions.

Discussion 61
Table 4.3
Summary of Results Related to Political Partisanship
Self-Rated
Political Ideology
Voted for Trump
(vs. Clinton)
Voted for
Someone Other
Than Trump or
Clinton Did Not Vote
Platforms used More
conservative
ideology
associated
with lower
likelihood of
getting news
from online
platforms
Less likely to get
news from print/
broadcast or
online platforms
More likely to
get news from
social media/
in-person
platforms
N/A
Overall
perception of
news reliability
N/A More likely to
believe the
news is less
reliable now
More likely to
believe the
news is less
reliable now
Less likely
to think the
news is more
reliable now
More-reliable
platforms
N/A More likely to
report the
most-reliable
news platforms
are cable
television and
social media
Less likely to
report that
print is the most
reliable news
platform
More likely to
report the
most-reliable
news platforms
are online and in
person
More likely
to report
the most
reliable news
platform is
social media
Gets news
from reliable
platforms
N/A N/A N/A N/A
Seeks out
differing views
More-
conservative
ideology
associated
with lower
likelihood of
“never or almost
never” seeking
out differing
views
More likely to
“always or
almost always”
or “never or
almost never”
seek out news
with differing
viewpoints
More likely
to “always or
almost always”
seek out news
with differing
viewpoints
N/A
NOTE: N/A indicates that variables were not significantly associated with the news
consumption profile factors in the regressions.

vision and social media as the most-reliable platforms, and they were
less likely to rate print as reliable. Compared with those who voted for
Clinton, those who voted for someone other than Trump or Clinton
were more likely to get news from social media/in-person, more likely
to believe the news is less reliable now than in the past, and more
likely to rate online and in-person platforms as most reliable. Taken
together, this suggests that those who voted for Clinton are less skepti-
cal of more-traditional news delivery platforms; they also tend to favor
print as their source for news, are much less likely to believe that the
reliability of news has declined over time, and less likely to rate online
or especially in-person news sources as the most-reliable platforms.
Those who voted for Clinton were also less likely than other voters to
report that they “always or almost always” seek out news sources that
offer views different from their own, in addition to the effect associated
with liberal ideology.
Factors Associated with News Consumption
Political Ideology and Behavior
The analyses presented in this report focus on demographic and politi-
cal factors linked to news consumption and find that both are associ-
ated with media use profiles and attitudes toward the platforms that
compose those profiles. Specifically, our results indicate that political
ideology and behavior are tied to news consumption—although politi-
cal behavior (voter turnout and choice) seems to be a stronger predic-
tor than partisanship. Self-rated political ideology and voting behavior
in the 2016 presidential election were each tied to different aspects of
news consumption and beliefs, but 2016 voting behavior appeared to
have a more substantively significant effect than partisanship and to
affect many more aspects of news consumption.
People who rated themselves as more conservative on the politi-
cal ideology scale scored significantly lower on the online dimension
of news consumption profiles than did more-liberal individuals, and
they were less likely to report that they “never or almost never” sought
out alternative viewpoints from the news. Those who self-identified as

Discussion 63
more liberal tended to score higher on the online consumption profile
and were more likely to report “never or almost never” seeking out
alternative views than more conservative respondents.
Voting behavior, in contrast, was associated with more aspects of
news consumption and beliefs than were underlying political ideolo-
gies. Our analyses report results relative to Clinton voters because they
were the largest category. First, people who reported voting for some-
one other than Clinton in 2016 were significantly (more than three
times) more likely than others to report a belief that the news is less
reliable now than in the past. Those who voted for third-party candi-
dates were more likely to rate online and in-person platforms as the
most reliable (four times more likely than Clinton voters) and to score
higher on the social media/in-person profile.
Trump voters were more likely than other voters to rate cable tele-
vision (two times) or social media (three and a half times) as the most-
reliable news platforms and significantly less likely to rate print as the
most reliable. They scored significantly lower than Clinton voters on
the print/broadcast dimension of news consumption profiles, and were
(three times) more likely than voters for any other candidate to report
either extreme (always or never) with regard to whether they sought
out news with alternative views to their own. Recall that a higher level
of self-identified conservative ideology was negatively associated with
reporting “never or almost never” seeking out news from differing
viewpoints.
Taken together, this suggests that those who voted for Clinton
rather than someone else were less skeptical of more-traditional news
delivery platforms; they tended to favor print as their source for news,
were much less likely to believe that the reliability of news has declined
over time, and were also less likely to rate online or especially in-person
news as the most-reliable platforms. Those who voted for Clinton were
also less likely than other voters to report that they “always or almost
always” seek out news sources that offer views different from their own.
Furthermore, those who did not vote in the 2016 presidential elec-
tion were even less likely than others to believe that the reliability of
news had improved over time, and, similar to Trump voters, were more
likely to rate social media as the most reliable news platform.

Demographics
It would be a mistake to suggest that individuals’ news consumption
profiles and their attitudes toward news reliability were influenced
only by political ideology or engagement, however.1 Instead, our results
underscored the importance of demographic factors—such as age, edu-
cation, and marital status.
Age. Age was shown to be an important predictor of news con-
sumption profile but of relatively little else. Younger individuals were
more likely to fit the social media/in-person and online consumption
profiles; older individuals were more likely to fit the print/broadcast
television profile. As previously noted, older individuals are more likely
to get news using more-traditional platforms than are younger people,
who are more likely to get news from social media/in-person or online.
Education. We observed educational differences in nearly every
topic we examined. Those with college educations were less likely
to report a heavy reliance on social media/in-person news compared
with radio and online sources. They were also less likely to think that
the reliability of the news had changed—for better or for worse—
compared with the past. College-educated individuals were more likely
to rate print and online platforms as being most reliable and were less
likely than those with lower levels of educational attainment to report
“never or almost never” seeking out news from alternative viewpoints.
Marital status. One demographic characteristic that appeared
several times in our analyses was the connection between marriage
and news consumption behaviors. Married people scored marginally
lower on their reliance on print/broadcast and were nearly three times
as likely as nonmarried individuals to rate in-person as the most reli-
able platform for news. Married people were also roughly one-half as
likely as others to report that they “always or almost always” sought
out sources with differing views. One possible explanation is that indi-
viduals cannot easily consume print media at the same time as other
1 Although political ideology and voting behavior were assessed several years before news
consumption habits, it is also possible that news consumption habits influence political
ideology and voting behaviors. To assess causal direction, we would require data tracking
changes in both over time.

Discussion 65
activities (such as housework) but can consume other platforms, such
as radio or in-person communication.
The prevailing theoretical framework for understanding the rela-
tionship between marriage and community life is that marriage is a
“greedy institution,” in that intense couple relationships come at the
expense of other connections and are particularly demanding of women’s
time through housework.2 Married women, but not men, are also signifi-
cantly less likely to volunteer after marriage and volunteer fewer hours.3
Obtaining news from platforms that respondents themselves have
rated as more reliable, or that offer a variety of viewpoints, might be a
similar premium that married people are less able to enjoy. This inter-
pretation is further strengthened by the finding that married people
who relied more heavily on social media/in-person platforms were
especially less likely to report getting news from platforms they viewed
as most reliable. At the same time, married people are significantly
more likely to vote in elections than unmarried people;4 those who
might encounter challenges to obtaining news and information from
platforms they deem most reliable, therefore, might also be among the
ones most engaged in political activity.
Implications for Truth Decay
These results have implications for how we might respond to Truth
Decay—the diminishing reliance on facts, data, and analysis in Ameri-
can public life—and its relationship to media and news consumption.
Our analyses suggest that news consumption choices might be influ-
2 Naomi Gerstel and Natalia Sarkisian, “Marriage: The Good, the Bad, and the Greedy,”
Contexts, Vol. 5, No. 4, 2006; and Natalia Sarkisian and Naomi Gerstel, “Till Marriage Do
Us Part: Adult Children’s Relationships with Their Parents,” Journal of Marriage and Family,
Vol. 70, No. 2, 2008.
3 Christopher J. Einolf and Deborah Philbrick, “Generous or Greedy Marriage? A Longitu-
dinal Study of Volunteering and Charitable Giving,” Journal of Marriage and Family, Vol. 76,
No. 3, 2014.
4 Jan E. Leighley and Jonathan Nagler, Who Votes Now? Demographics, Issues, Inequality,
and Turnout in the United States, Princeton, N.J.: Princeton University Press, 2013.

enced not only by political beliefs but also by demographics, respon-
sibilities, and time demands or constraints. If this is true, then just
improving the quality of information available or providing individuals
skills to understand and evaluate media sources more effectively might
not be sufficient to address Truth Decay. Instead, doing so might
require changing people’s attitudes toward consuming news (reinforc-
ing its importance and clarifying its benefits) and offering new ways to
access reliable news that are as easy as social media and in-person com-
munication. What those new platforms or mechanisms might look like
should be an area for future research.
More broadly, this report’s analyses present three key implications
relevant to the overall Truth Decay agenda, a part of which is focused
on understanding how people use and relate with news.
Consumption Choices and Social Media
Much attention has been paid to the increasing reliance of news con-
sumers on social media as a source of news because of the ease with
which misinformation and disinformation spreads through that plat-
form and because of perceptions that the quality of its information
and discourse is lower than that on other platforms.5 We do not evalu-
ate these arguments in this report, but we can offer some insight into
trends in the use of social media for news.
According to our analysis, 30 percent of respondents reported
relying on social media as their primary or secondary source of news,
ranking it somewhere in the middle in terms of use compared with
other sources. However, there is evidence in our data that even those
30 percent of consumers who do rely heavily on social media for news
are likely also using other sources of news. This is consistent with past
work discussed at the beginning of this report focused on media com-
plementarity and the way that different sources might function together
to meet an individual’s information needs.6 Continuing to track how
5 Hunt Alcott, Matthew Gentzkow, and Chuan Yu, Trends in the Diffusion of Misinformation
on Social Media, Cambridge, Mass.: National Bureau of Economic Research, NBER Working
Paper No. 25500, 2018; and Rande Price, “Oxford Research Documents the Alarming Rise of
Disinformation on Social Media,” Digital Content Next, September 21, 2018.
6 See for example, Van Damme et al., 2015.

Discussion 67
people use social media—and especially which specific sources they
rely on—will be an important area for future research.
These findings are relevant to a recent study by Grinberg and col-
leagues that estimated that exposure to fake news on Twitter during
the 2016 U.S. presidential election was rare, but that engagement with
fake news sources was extremely concentrated: 1 percent of individu-
als accounted for 80 percent of fake news exposures, and 0.1 percent
accounted for nearly 80 percent of fake news sources shared.7 Likeli-
hood of engaging with fake news sources on Twitter increased sub-
stantially for older people and exponentially for political conservatives.
Our results show that those who voted for Trump are much more likely
to rate social media as the most reliable platform, but conservatives in
general are more likely to think that the news is less reliable now. Thus,
one interpretation of their overall perception of reduced news reliabil-
ity is that it might be a response to greater exposure to fake news.
However, their likelihood of using social media platforms is not dif-
ferent from more-liberal respondents, and younger people overall were
more likely to rely on social media platforms, which raises questions
about whether there are further differences among individuals in their
news-accessing and news-sharing behaviors on social media. We plan
to examine this question in future research.
News Consumption and Demographics
Our results indicate that people obtain news in different ways—
relying on different types of sources and with varying degrees of trust
in the news overall and in specific news delivery platforms. Partisan-
ship explains some of these differences but certainly not all of them.
Such factors as vote choice, propensity to vote, age, gender, education,
and marital status all play a significant role in explaining news con-
sumption. Different news consumption profiles are associated with
different personal and demographic attributes and living situations and
so might also reflect an individual’s access to information in terms of
time, technology, and availability.
7 Nir Grinberg, Kenneth Joseph, Lisa Friedland, Briony-Swire-Thompson, and David Lazer,
“Fake News on Twitter During the 2016 U.S. Presidential Election,” Science, Vol. 363, No.
6425, 2019.

The results in this report consider choices between platforms, not
specific news outlets, but these insights still have broad implications for
efforts aimed at improving how people access, interpret, and evaluate
news. Although much attention has been paid to media literacy educa-
tion as a necessity in the information environment, the analysis here
suggests that perceptions of reliability and issues surrounding ease of
access to reliable news might be equally important.
Media Reliability
Finally, this study considered perceptions of media reliability and trust in
the media. Our analysis confirmed that many people are skeptical of the
reliability of news overall but that a small minority thinks that reliability
is increasing. This underscores the importance of considering disaggre-
gated data when studying media consumption and trust in the media.
Perceptions of the overall reliability of the news are linked to
whether people seek out alternative viewpoints in the news they con-
sume. People who believe the news to be more reliable now than in the
past are more likely to report that they “always or almost always” seek
out alternative views in the news, and people who report any change in
the overall reliability of the news (either an increase or a decrease in reli-
ability) are less likely than those who think the news is as reliable now as
in the past to report that they “never or almost never” seek out alternative
views. Furthermore, individuals who say that the news is less reliable now
than in the past are about one-third less likely than others to primarily
obtain news from platforms they rate as the most reliable. Rather than
rely more heavily on news platforms they believe to be the most reli-
able as a means to mitigate perceived decreases in overall news reliability,
these individuals are at greater risk of doing the contrary.
Attitudes also vary about the reliability of different types of news
platforms. Broadcast and cable news are considered the most-reliable
sources, followed by print and online sources, which in turn are fol-
lowed by social media and in-person communication. Our analysis also
shows that attitudes toward the media (whether measured as reliability
or trust) are linked to consumption. For most users, perceived reliabil-
ity matters. News media consumers tend to rely most heavily on those
sources that they perceive as most reliable, though, as we have noted,

Discussion 69
there are some exceptions, such as those who rely on other people and
social media for news, some married people, and (perhaps surprisingly)
those who believe the news is less reliable overall. Clearly the belief that
news is less reliable now than in the past is not sufficient to motivate
many people to seek out news that they do believe comes from reliable
platforms—in fact, our findings are the opposite.
Prior studies have tended to use the terms trust and credibility
interchangeably; our use of the word reliability was intended to assess
the degree to which respondents felt that they could rely or consistently
depend on a given news platform to provide accurate information.
Nonetheless, we were able to show that direct assessment of “trust”
in various news platforms mapped strongly onto the reliability-based
results, indicating a close similarity to the literature on media trust.
Much attention has been paid to the declining trust in media
organizations and the effect of that decline on the ability of the media
to serve in its role as provider of information and on the willingness
of individuals to engage seriously with news.8 The results here, how-
ever, remind us that factors related to both supply and demand could
improve the trust that individuals have in the media. For some con-
sumers, trust and consumption appear to be closely linked, so that
levels of trust in media might drive consumption choices, and con-
sumption choices, in turn, might feed back into trust. For others, trust
might not be as closely linked to consumption but might be shaped
by environmental or individual characteristics. Future work examining
this potential feedback loop over time is warranted. For all consum-
ers, choices about media consumption are likely linked to individual
characteristics, preferences, and behaviors. As a result, attempts to alter
or improve the relationship between media outlets and consumers are
likely to require attention not only to the supply side of the information
ecosystem but also to the demand side and the ways in which people
access news. Future work that explores news consumption in this more
holistic way will be important as the informational ecosystem grows
increasingly diverse and complex.
8 For example, see Chris Peters and Marcel Jeroen Broersma, eds., Rethinking Journalism:
Trust and Participation in a Transformed News Landscape, New York: Routledge, 2013.

71
APPENDIX A
Ranked Data Factor Analysis Procedure
This appendix describes the factor analysis procedure that was used to
identify the news consumption profiles.
Method
We used EFA to analyze the platforms that people use to receive news,
focusing on print, broadcast, cable, radio, online, social media, or in-
person communication.
The participants were asked to rank the top four out of the seven
media sources. We coded the top four as 1–4, with 4 being the high-
est rank. We treated the rest of the unranked sources (i.e., missing
responses) as rank 0.
We then proceeded with a monotonic transformation of ranked
sources via the PROC PRINQUAL procedure followed by EFA in
SAS. Our initial principal components analysis suggests four princi-
pal components. Using that as a guideline, the PROC PRINQUAL
procedure transforms each variable to be as much as possible like the
first principal component (or, more generally, to be close to the space
defined by the first N = 4 principal components). Before PRINQUAL,
four components account for 70 percent of the variance of the untrans-
formed ranked data. After the monotonic transformation, more than
99 percent of the variance of the transformed data were explained by
the four extracted factors (see Figure A.1).
With the transformed responses, we proceed with EFA to inves-
tigate types of communication using the VARIMAX orthogonal rota-

tion. Eigenvalues,1 scree plot,2 and proportion of variance explained
were used to guide decisions about the number of factors to extract
from seven sources of information. Based on the rotated factor load-
ing pattern, we examined the interpretability of EFA results. When the
criteria of the number of factors extracted were satisfied and the results
were interpretable, we proceeded to scoring, using the coefficients gen-
erated from EFA results.
Results
EFA results indicated a four-factor solution based on an overall evalua-
tion of eigenvalues, scree plot, and variance explained. Kaiser’s criterion
considers factors with an eigenvalue greater than 1 as common factors.
As is shown in the scree plot (see Figure A.1), four eigenvalues are above
1 and the biggest drop in eigenvalues is between #4 and #5, suggest-
1 Henry F. Kaiser, “The Application of Electronic Computers for Factor Analysis,” Educa-
tional and Psychological Measurement, Vol. 20, No. 1, 1960.
2 Raymond B. Cattell, “The Scree Test for the Number of Factors,” Multivariate Behavioral
Research, Vol. 1, No. 2, 1966.
Figure A.1
Scree Plot and Variance Explained by Factor Analysis, Suggesting Four
Factors Should Be Retained
Eigenvalue
2.5
2.0
1.5
1.0
0.5
0
654321 7
Factor
Eigenvalue
2.5
2.0
1.5
1.0
0.5
0
654321 7
Factor
Scree Plot Variance Explained
Cumulative
Proportion

Ranked Data Factor Analysis Procedures 73
ing four factors be kept. The loading patterns (see Table A.1) suggest
four primary ways in which people received information: (1) Social
media and in-person communication were highly ranked; (2) print and
broadcast television were highly ranked; (3) online was highly ranked;
and (4) radio was highly ranked. Scoring coefficients (also shown in
Table A.1) are used to generate factor scores for each person on each of
these four dimensions. Scores on each factor have a mean of 0 and a
standard deviation of 1.
Table A.1
EFA Estimated Loadings and Scoring Coefficients
Factor 1 Factor 2 Factor 3 Factor 4
EFA estimated factor loadings
Print –0.02 0.98 0.02 0.20
Broadcast 0.00 0.83 0.00 –0.55
Cable –0.57 –0.43 –0.69 –0.09
Radio –0.02 0.02 0.02 1.00
Online –0.18 –0.09 0.98 0.00
Social media 1.00 –0.02 –0.04 –0.02
In person 1.00 –0.02 –0.04 –0.02
Scoring coefficients
Print –0.08 0.68 –0.13 1.02
Broadcast –0.02 0.29 0.02 –1.38
Cable –0.23 –0.19 –0.41 –0.28
Radio 0.00 0.00 0.00 0.00
Online –0.12 –0.15 0.73 –0.23
Social media 0.85 –0.12 –0.11 –0.18
In person 0.00 0.00 0.00 0.00

75
APPENDIX B
Study Descriptive Statistics
Table B.1 lists the unweighted statistics for the variables used in this
report.

Table B.1
Study Variable Descriptive Statistics (Unweighted)
Variable
Mean or
Proportion
Standard
Deviation
Age 56.66 14.30
Sex
Male 0.44 0.50
Female (reference) 0.56
Race/ethnicity
Non-Hispanic black 0.10 0.29
Hispanic 0.14 0.35
Other race 0.05 0.22
Non-Hispanic white (reference) 0.71 0.45
Has a college degree (yes) 0.23 0.42
Household income (in thousands) 76.17 60.40
Married (yes) 0.58 0.49
Any children in the household (yes) 0.47 0.50
Retired (yes) 0.32 0.46
Do you find the news you currently receive . . .
. . . Less reliable than in the past 0.41 0.49
. . . More reliable than in the past 0.15 0.36
. . . About the same (reference) 0.44
News consumption profile factors
Social media/in-person 0.00 1.00
Print/broadcast 0.00 1.00
Radio 0.00 1.00
Online 0.00 1.00
Partisanship
Conservatism
“How would you describe your political
viewpoints?” (on a scale from 1 = very liberal to
5 = very conservative)
2.96 1.16
Voted for Trump 0.36 0.48
Voted for someone other than Trump or Clinton 0.08 0.27
Did not vote 0.09 0.29
Voted Clinton (reference) 0.47 0.50

77
APPENDIX C
Trust in Institutions and News Consumption
As a supplement to our analysis of how perceived reliability shapes
news consumption, we explored the link between perceived reliability
of news platforms and explicit trust in particular news platforms to
further validate our reliability measure. We also measured the links
between trust and news consumption profiles. To explore how trust
affects news consumption, we linked the data for our 2018 survey
with data on institutional trust collected from a separate survey of
1,009 ALP respondents conducted in April 2018 by matching the
data for common participants. The survey contained attitudes toward
national and local newspapers, cable and broadcast television, and
social media, ranging from complete distrust to complete trust.1
We then conducted a multinomial regression using reported trust
in news platforms (from April 2018) as an explanatory variable and
perceived reliability as the dependent variable. In short, the results are
similar to the previous results in Table 3.6 but factor in the trust mea-
sures as additional covariates. We find that reported trust and per-
ceived reliability are closely correlated. Trust in national newspapers is
strongly predictive of rating print as the most reliable news platform;
trust in cable news is similarly linked to rating cable television as the
most reliable platform; trust in broadcast news is significantly linked
with rating broadcast television as more reliable than cable television,
radio, or online news platforms. Finally, trust in social media was sig-
1 All respondents in the institutional trust survey were participants in the larger media
reliability survey. We were able to match 38 percent of the responses in the original data set
using the new data.

nificantly linked to rating social media as the more reliable platform
and to rating cable television and print as more reliable than broadcast
television. Thus, our results suggest that, conceptually, trust and reli-
ability are closely related for consumers when assessing various news
platforms.
The measures of trust similarly predicted the use patterns iden-
tified in the media consumption profiles. Specifically, trust in social
media is associated with a higher likelihood of being in the social media/
in-person group (so, those who report higher trust in social media are
more likely to use social media and in-person news sources most often).
Trust in national newspapers was positively associated with being in
the print/broadcast group. Trust in cable news is negatively associated
with being in the social media/in-person, print/broadcast, and online
consumption profiles, and trust in broadcast news is negatively asso-
ciated with the likelihood of being in the radio consumption profile.
These results suggest that just as perceived reliability is associated with
news consumption choices, a respondent’s trust in various news plat-
forms is also associated with information use habits.
Table C.1 reports the results of a multinomial regression using
reported trust in news platforms (from April 2018) as an explanatory
variable and perceived reliability as the dependent variable. The results
in Table C.1 are similar to those in Table 3.6 but factor in the trust
measures as additional covariates.
Table C.2 is similar to the OLS regressions linking characteristics
to news consumption profiles in Table 3.1 but add in the trust mea-
sures as further controls.

Trust in Institutions and News Consumption 79
Table C.1
Most Reliable News Platform, Compared with Broadcast Television,
Including Trust Measures (Relative Risk Ratios)
Characteristics Print
Cable
Television Radio Online
Social
Media In-Person
Age 0.993 1.003 0.976 0.967* 0.952† 0.984
Male 2.005* 1.298 3.593** 2.487* 0.380 1.359
Black 0.167* 0.425 0.100* 0.516 0.000*** 0.174
Hispanic 1.378 1.005 0.051*** 0.742 0.725 1.062
Other race 1.248 1.423 6.998* 3.600 6.113 0.000***
College 1.258 0.700 0.555 1.849 6.343 4.017*
Income 1.007* 0.998 1.005 0.996 0.969* 0.975**
Married 0.993 1.213 0.803 1.829 0.670 2.599
Any children 0.386** 1.008 1.086 0.772 0.483 1.361
Retired 1.418 1.645 1.090 2.025 6.134† 2.193
Conservatism 0.934 0.724† 0.716 0.768 3.492** 1.006
Voted for
Trump
0.458 0.969 1.151 2.339 1.346 0.421
Voted for
someone
other than
Trump or
Clinton
1.615 0.780 1.600 9.336*** 2.581 1.440
Did not vote 0.098* 0.367† 0.266 1.003 7.186† 0.273
Trust in
national
newspapers
1.238* 0.741** 1.024 1.356* 0.745 0.695**
Trust in local
newspapers
0.996 0.923 1.011 0.897 0.783 0.937
Trust in cable
news
0.934 1.517*** 1.084 1.053 0.780 0.954
Trust in
broadcast news
0.862 0.809* 0.760* 0.645** 1.150 1.009
Trust in social
media
1.092 1.147† 0.878 1.090 2.546*** 1.064
NOTE: For this analysis, N = 1,008. † p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001.

Table C.2
Characteristics Associated with News Consumption Profile Factors,
Including Trust Measures (OLS Coefficients)
Social Media/
In-Person
Print/
Age –0.016*** 0.008† –0.005 –0.014**
Male –0.240** –0.060 0.115 0.070
Black –0.065 0.331* 0.039 –0.239
Hispanic –0.200 –0.199 –0.423* –0.131
Other race –0.474 0.022 0.053 0.564
College –0.073 0.063 0.141† 0.093
Income –0.002* 0.000 0.000 0.001
Married –0.205* –0.020 0.017 0.015
Any children 0.126 –0.151 –0.030 0.025
Retired –0.052 –0.129 –0.171 0.102
Conservatism –0.027 0.088 –0.017 –0.008
Voted for Trump 0.215 0.046 0.031 –0.295†
Voted for someone
other than Trump or
Clinton
0.535** –0.373 –0.070 0.074
Did not vote 0.105 –0.025 –0.032 –0.110
Trust in national
newspapers
0.014 0.099*** 0.041 –0.006
Trust in local
newspapers
–0.005 –0.031 –0.031 0.006
Trust in cable news –0.090** –0.045† 0.005 –0.050†
Trust in broadcast
news
0.036 0.026 –0.090* 0.002
Trust in social media 0.052* –0.041 –0.010 –0.004
NOTE: For this analysis, N = 1,008. † p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001.

81
References
Ahlers, Douglas, “News Consumption and the New Electronic Media,” Harvard
International Journal of Press/Politics, Vol. 11, No. 1, 2006, pp. 29–52.
Alcott, Hunt, Matthew Gentzkow, and Chuan Yu, Trends in the Diffusion of
Misinformation on Social Media, Cambridge, Mass.: National Bureau of Economic
Research, NBER Working Paper No. 25500, 2018.
Althaus, Scott, Anne M. Cizmar, and James G. Gimpel, “Media Supply, Audience
Demand, and the Geography of News Consumption in the United States,”
Political Communication, Vol. 26, No. 3, 2009, pp. 249–277.
Baldwin, Thomas F., and Marianne Barrett, “Uses and Values for News on Cable
Television,” Journal of Broadcasting and Electronic Media, Vol. 36, No. 2, 1992,
pp. 225–234.
Barthel, Michael, “Newspapers Fact Sheet,” webpage, Pew Research Center,
June 13, 2018. As of August 4, 2018:
http://www.journalism.org/fact-sheet/newspapers/
Bennet, Stephen Earl, Staci L. Rhine, and Richard S. Flickinger, “Assessing
Americans’ Opinions About the News Media’s Fairness in 1996 and 1998,”
Political Communication, Vol. 18, No. 2, 2001, pp. 163–182.
Bennett, Stephen Earl, Staci L. Rhine, Richard S. Flickinger, and Linda
L. M. Bennett, “‘Video Malaise’ Revisited: Public Trust in the Media and
Government,” Harvard International Journal of Press/Politics, Vol. 4, No. 4, 1999,
pp. 8–23.
Blumler, Jay G., and Elihu Katz, The Uses of Mass Communications, Beverly Hills,
Calif.: Sage Publications, 1974.
Bogart, Leo, “The Public’s Use and Perception of Newspapers,” Public Opinion
Quarterly, Vol. 48, No. 4, January 1, 1984, pp. 709–719.
Bucy, Erik P., “Media Credibility Reconsidered: Synergy Effects Between On-Air
and Online News,” Journalism and Mass Communication Quarterly, Vol. 80, No. 2,
2003, pp. 247–264.

Carman, Katherine, “Well Being 496: Trust in Government,” RAND American
Life Panel, webpage, April 2018. As of September 5, 2019:
https://alpdata.rand.org/index.php?page=data&p=showsurvey&syid=496
Carter, Richard F., and Bradley S. Greenberg, “Newspaper or Television: Which
Do You Believe?” Journalism Quarterly, Vol. 42, No. 1, 1965, pp. 29–34.
Cattell, Raymond B., “The Scree Test for the Number of Factors,” Multivariate
Behavioral Research, Vol. 1, No. 2, 1966, pp. 245–276.
Chan, Joey Ka-Ching, and Louis Leung, “Lifestyles, Reliance on Traditional
News Media and Online News Adoption,” New Media & Society, Vol. 7, No. 3,
2005, pp. 357–382.
Coffey, Steve, and Horst Stipp, “The Interactions Between Computer and
Television Usage,” Journal of Advertising Research, Vol. 37, No. 2, 1997, pp. 61–68.
Dutta-Bergman, Mohan J., “Complementarity in Consumption of News Types
Across Traditional and New Media,” Journal of Broadcasting & Electronic Media,
Vol. 48, No. 1, 2004, pp. 41–60.
Einolf, Christopher J., and Deborah Philbrick, “Generous or Greedy Marriage? A
Longitudinal Study of Volunteering and Charitable Giving,” Journal of Marriage
and Family, Vol. 76, No. 3, 2014, pp. 573–586.
Faber, Ronald J., Stephen D. Reese, and Leslie H. Steeves, “Spending Time
with the News Media: The Relationship Between Reliance and Use,” Journal of
Broadcasting and Electronic Media, Vol. 29, No. 4, 1985, pp. 445–450.
Fischer, Sara, “92% of Republicans Think Media Intentionally Reports Fake
News,” Axios, June 27, 2018. As of August 4, 2018:
https://www.axios.com/trump-effect-92-percent-republicans-media-fake-news-
9c1bbf70-0054-41dd-b506-0869bb10f08c.html
Flanagin, Andrew J., and Miriam J. Metzger, “Internet Use in the Contemporary
Media Environment,” Human Communication Research, Vol. 27, No. 1, 2001,
pp. 153–181.
Fletcher, Richard, and Sora Park, “The Impact of Trust in the News Media on
Online News Consumption and Participation,” Digital Journalism, Vol. 5, No. 10,
2017, pp. 1281–1299.
Garrett, R. Kelly, “Echo Chambers Online? Politically Motivated Selective
Exposure Among Internet News Users,” Journal of Computer-Mediated
Communication, Vol. 14, No. 2, 2009, pp. 265–285.
Garrett, R. Kelly, and Natalie Jomini Stroud, “Partisan Paths to
Exposure Diversity: Differences in Pro- and Counterattitudinal News
Consumption,” Journal of Communication, Vol. 64, No. 4, 2014, pp. 680–701.

References 83
Gaskins, Benjamin, and Jennifer Jerit, “Internet News: Is It a Replacement for
Traditional Media Outlets?” International Journal of Press/Politics, Vol. 17, No. 2,
2012, pp 190–213.
Gaziano, Cecilie, and Kristin McGrath, “Measuring the Concept of Credibility,”
Journalism Quarterly, Vol. 63, No. 3, 1986, pp. 451–462.
Gentzkow, Matthew, “Valuing New Goods in a Model with Complementarity:
Online Newspapers,” American Economic Review, Vol. 97, No. 3, 2007,
pp. 713–744.
Gerstel, Naomi, and Natalia Sarkisian, “Marriage: The Good, the Bad, and the
Greedy,” Contexts, Vol. 5, No. 4, 2006, pp. 16–21.
Glynn, Carroll J., Michael E. Huge, and Lindsay H. Hoffman, “All the News
That’s Fit to Post: A Profile of News Use on Social Networking Sites,” Computers
in Human Behavior, Vol. 28, No. 1, 2012, pp. 113–119.
Grinberg, Nir, Kenneth Joseph, Lisa Friedland, Briony-Swire-Thompson, and
David Lazer, “Fake News on Twitter During the 2016 U.S. Presidential Election,”
Science, Vol. 363, No. 6425, 2019, pp. 374–378.
Gunther, Albert C., “Biased Press or Biased Public?” Public Opinion Quarterly,
Vol. 56, No. 2, 1992, pp. 147–167.
Haller, Max, Markus Hadler, and Gerd Kaup, “Leisure Time in Modern Societies:
A New Source of Boredom and Stress?” Social Indicators Research, Vol. 111, No. 2,
2013, pp. 403–434.
Hart, William, Dolores Albarracín, Alice H. Eagly, Inge Brechan, Matthew J.
Lindberg, and Lisa Merrill, “Feeling Validated Versus Being Correct: A Meta-
Analysis of Selective Exposure to Information,” Psychological Bulletin, Vol. 135,
No. 4, 2009, pp. 555–588.
Hastall, Matthias R., “Information Utility as Determinant of Media Choices,” in
Tilo Hartmann, ed., Media Choice: A Theoretical and Empirical Overview, New
York: Routledge, 2009, pp. 149–166.
Hawkins, Robert P., Suzanne Pingree, Jacqueline Hitchon, Bradley W. Gorham,
Prathna Kannaovakun, Eileen Gilligan, Barry Radler, Gudbjorg H. Kolbeins, and
Toni Schmidt, “Predicting Selection and Activity in Television Genre Viewing,”
Media Psychology, Vol. 3, No. 3, 2001, pp. 237–264.
Hmielowski, Jay D., R. Lance Holbert, and Jayeon Lee, “Predicting the
Consumption of Political TV Satire: Affinity for Political Humor, The Daily
Show, and The Colbert Report,” Communication Monographs, Vol. 78, No. 1, 2011,
pp. 96–114.
Huang, Edgar, “The Causes of Youths’ Low News Consumption and Strategies for
Making Youths Happy News Consumers,” Convergence: The International Journal
of Research into New Media Technologies, Vol. 15, No. 1, 2009, pp. 105–122.

Iyengar, Shanto, and Kyu S. Hahn, “Red Media, Blue Media: Evidence of
Ideological Selectivity in Media Use,” Journal of Communication, Vol. 5, 2009,
pp. 19–39.
Jakob, Nikolaus G., “No Alternatives? The Relationship Between Perceived Media
Dependency, Use of Alternative Information Sources, and General Trust in Mass
Media,” International Journal of Communication, Vol. 4, 2010, pp. 589–606.
Johnson, Thomas J., and Barbara K. Kaye, “Cruising Is Believing? Comparing
Internet and Traditional Sources on Media Credibility Measures,” Journalism and
Mass Communication Quarterly, Vol. 75, No. 2, 1998, pp. 325–340.
Jones, David A., “Why Americans Don’t Trust the Media: A Preliminary
Analysis,” Harvard International Journal of Press/Politics, Vol. 9, No. 2, 2004,
pp. 60–75.
Kaiser, Henry F., “The Application of Electronic Computers for Factor Analysis,”
Educational and Psychological Measurement, Vol. 20, No. 1, 1960, pp. 141–151.
Kavanagh, Jennifer, and Michael D. Rich, Truth Decay: An Initial Exploration of
the Diminishing Role of Facts and Analysis in American Public Life, RR-2314-RC,
2018. As of September 3, 2019:
https://www.rand.org/pubs/research_reports/RR2314.html
Kayany, Joseph M., and Paul Yelsma, “Displacement Effects of Online Media in
the Socio-Technical Contexts of Households,” Journal of Broadcasting & Electronic
Media, Vol. 44, No. 2, 2000, pp. 215–229.
Kaye, Barbara K., and Thomas J. Johnson, “From Here to Obscurity? Media
Substitution Theory and Traditional Media in an Online World,” Journal of the
American Society for Information Science and Technology, Vol. 54, No. 3, 2003,
pp. 260–273.
Kiousis, Spiro, “Public Trust of Mistrust? Perceptions of Media Credibility
in the Information Age,” Mass Communication & Society, Vol. 4, No. 4, 2001,
pp. 381–403.
Knight Foundation, An Online Experimental Platform to Assess Trust in the Media,
Miami, Fla., July 18, 2018. As of August 4, 2018:
https://knightfoundation.org/reports/
an-online-experimental-platform-to-assess-trust-in-the-media
Knight Foundation and Trust, Media and Democracy, “10 Reasons Why
American Trust in the Media Is at an All-Time Low,” Medium, January 15, 2018.
As of August 4, 2018:
https://medium.com/trust-media-and-democracy/10-reasons-why-americans-dont-
trust-the-media-d0630c125b9e
Kohring, Matthias, and Jörg Matthes, “Trust in News Media: Development and
Validation of a Multidimensional Scale,” Communication Research, Vol. 34, No. 2,
2007, pp. 231–252.

References 85
Krcmar, Marina, and Yuliya Strizhakova, “Uses and Gratifications as Media
Choice,” in Tilo Hartmann, ed., Media Choice: A Theoretical and Empirical
Overview, New York: Routledge, 2009, pp. 53–69.
Lavrakas, Paul, “Response Bias,” Encyclopedia of Survey Research Methods,
Thousand Oaks, Calif.: Sage, 2008.
Lee, Angela M., “News Audiences Revisited: Theorizing the Link Between
Audience Motivations and News Consumption,” Journal of Broadcasting &
Electronic Media, Vol. 57, No. 3, 2013, pp. 300–317.
Lee, Tien-Tsun, “Why They Don’t Trust the Media: An Examination of Factors
Predicting Trust,” American Behavioral Scientist, Vol. 54, No. 1, 2010, pp. 8–21.
Leighley, Jan E., and Jonathan Nagler, Who Votes Now? Demographics, Issues,
Inequality, and Turnout in the United States, Princeton, N.J.: Princeton University
Press, 2013.
Liebler, Carolyn A., and Gary D. Sandefur, “Gender Differences in the Exchange
of Social Support with Friends, Neighbors, and Co-Workers at Midlife,” Social
Science Research, Vol. 31, No. 3, 2002, pp. 364–391.
Liebowitz, Stan J., and Alejandro Zentner, “Clash of the Titans: Does Internet Use
Reduce Television Viewing?” Review of Economics and Statistics, Vol. 94, No. 1,
2012, pp. 234–245.
Lin, Carolyn, “Modeling the Gratification-Seeking Process of Television Viewing,”
Human Communication Research, Vol. 20, No. 2, 1993, pp. 224–244.
Lin, Carolyn, Michael B. Salwen, Bruce Garrison, and Paul D. Driscoll, “Online
News as a Functional Substitute for Offline News,” in Michael B. Salwen, Bruce
Garrison, and Paul D. Driscoll, eds., Online News and the Public, Mahwah, N.J.:
Lawrence Erlbaum, 2005, pp. 237–256.
Mackay, Jenn Burleson, and Wilson Lowrey, “The Credibility Divide: Reader
Trust of Online Newspapers and Blogs,” Journal of Media Sociology, Vol. 3,
Nos. 1–4, 2011, pp. 39–57.
Martin, Vivian B., “Attending the News: A Grounded Theory About a Daily
Regimen,” Journalism, Vol. 9, No. 1, 2008, pp. 76–94.
Mason, Liliana, Uncivil Agreement: How Politics Became Our Identity, Chicago, Ill.:
University of Chicago Press, 2018.
Matsa, Katerina Eva, and Elisa Shearer, “News Use Across Social Media Platforms
2018,” webpage, Pew Research Center, September 10, 2018. As of November 4,
2018:
http://www.journalism.org/2018/09/10/
news-use-across-social-media-platforms-2018/

McDonald, Steve, and Christine A. Mair, “Social Capital Across the Life Course:
Age and Gendered Patterns of Network Resources,” Sociological Forum, Vol. 25,
No. 2, June 2010.
Merriam-Webster, “Filter Bubble,” webpage, undated. As of August 4, 2018:
https://www.merriam-webster.com/dictionary/filter%20bubbles
Meyer, Philip, “Defining and Measuring Credibility of Newspapers: Developing
an Index,” Journalism Quarterly, Vol. 65, No. 3, 1988, pp. 567–588.
Mitchell, Amy, Jeffrey Gottfried, Elisa Shearer, and Kristine Lu, How Americans
Encounter, Recall and Act Upon Digital News, Washington, D.C.: Pew Research
Center, February 9, 2017. As of August 4, 2018:
how-americans-encounter-recall-and-act-upon-digital-news/
Mitchell, Amy, Katie Simmons, Katerina Eva Matsa, and Laura Silver, “Publics
Globally Want Unbiased News Coverage, but Are Divided on Whether Their
News Media Deliver,” Pew Research Center, January 11, 2018. As of August 4,
2018:
http://www.pewglobal.org/2018/01/11/publics-globally-want-unbiased-news-
coverage-but-are-divided-on-whether-their-news-media-deliver/
Mitchelstein, Eugenia, and Pablo J. Boczkowski, “Online News Consumption
Research: An Assessment of Past Work and an Agenda for the Future,” New Media
& Society, Vol. 12, No. 7, 2010, pp. 1085–1102.
Newhagen, John, and Clifford Nass, “Differential Criteria for Evaluating
Credibility of Newspapers and TV News,” Journalism Quarterly, Vol. 66, No. 2,
1989, pp. 277–284.
Owen, Laura Hazard, “Here’s Who Gets News from TV: The Elderly, Pew Finds
(Again),” Neiman Lab, January 8, 2018. As of August 4, 2018:
http://www.niemanlab.org/2018/01/
heres-who-gets-news-from-tv-the-elderly-pew-finds-again/
Papacharissi, Zizi, and Alan M. Rubin, “Predictors of Internet Use,” Journal of
Broadcasting & Electronic Media, Vol. 44, No. 2, 2000, pp. 175–196.
Peters, Chris, “Journalism to Go,” Journalism Studies, Vol. 13, Nos. 5–6, 2012,
pp. 695–705.
Peters, Chris, and Marcel Jeroen Broersma, eds., Rethinking Journalism: Trust and
Participation in a Transformed News Landscape, New York: Routledge, 2013.
Pew Research Center, One-in-Ten Voters Online for Campaign ’96, Washington,
D.C., December 16, 1996. As of August 4, 2018:
http://assets.pewresearch.org/wp-content/uploads/sites/5/legacy-pdf/117.pdf

References 87
Pollard, Michael, and Matthew D. Baird, The RAND American Life Panel:
Technical Description, Santa Monica, Calif.: RAND Corporation, RR-1651, 2017.
As of September 3, 2019:
Pollard, Michael, and Joshua Mendelsohn, “RAND Kicks Off 2016 Presidential
Election Panel Survey,” The RAND Blog, January 27, 2016. As of August 4, 2018:
https://www.rand.org/blog/2016/01/rand-kicks-off-2016-presidential-election-
panel-survey.html
———, Methodology of the 2016 Rand Presidential Election Panel Survey (PEPS),
Santa Monica, Calif.: RAND Corporation, RR-1460-RC/UCLA, 2016. As of
September 3, 2019:
Price, Rande, “Oxford Research Documents the Alarming Rise of Disinformation
on Social Media,” Digital Content Next, September 21, 2018.
Prior, Markus, “News vs. Entertainment: How Increasing Media Choice Widens
Gaps in Political Knowledge and Turnout,” American Journal of Political Science,
Vol. 49, No. 3, July 2005, pp. 577–592.
RAND Corporation, “Welcome to the ALP Data Pages,” webpage, undated. As of
September 5, 2019:
https://alpdata.rand.org
Rimmer, Tony, and David Weaver, “Different Questions, Different Answers?
Media Use and Media Credibility,” Journalism Quarterly , Vol. 64, No. 1, 1987,
pp. 28–44.
Rubin, Alan M., “Uses and Gratifications: An Evolving Perspective of Media
Effects,” in Robin L. Nabi and Mary Beth Oliver, eds., The SAGE Handbook of
Media Processes and Effects, Thousand Oaks, Calif.: SAGE Publications, 2009,
pp. 147–159.
Ruggiero, Thomas E., “Uses and Gratifications Theory in the 21st Century,” Mass
Communication & Society, Vol. 3, No. 1, 2001, pp. 3–37.
Sarkisian, Natalia, and Naomi Gerstel, “Till Marriage Do Us Part: Adult
Children’s Relationships with Their Parents,” Journal of Marriage and Family,
Vol. 70, No. 2, 2008, pp. 360–376.
Schmidt, Christine, “Planted Stories? Fake News as Editorial Decisions? Trump or
CNN? A Poll Examines the Public’s Trust of Mainstream Media,” Neiman Lab,
April 3, 2018. As of August 4, 2018:
http://www.niemanlab.org/2018/04/planted-stories-fake-news-as-editorial-
decisions-trump-or-cnn-a-poll-examines-the-publics-trust-of-mainstream-media/
Shearer, Elisa, “Audio and Podcasting Fact Sheet,” webpage, Pew Research Center,
July 12, 2018. As of August 4, 2018:
http://www.journalism.org/fact-sheet/audio-and-podcasting/

Shearer, Elisa, and Jeffrey Gottfried, “News Use Across Social Media Platforms
2017,” webpage, Pew Research Center, September 7, 2017. As of August 4, 2018:
news-use-across-social-media-platforms-2017/
Shoemaker, Pamela J., “Hardwired for News: Using Biological and Cultural
Evolution to Explain Surveillance Function,” Journal of Communication, Vol. 46,
No. 3, 1996, pp. 32–47.
Stempel, Guido H., and Thomas Hargrove, “Mass Media Audiences in a
Changing Media Environment,” Journalism & Mass Communication Quarterly,
Vol. 73, No. 3, September 1996, pp. 549–558.
Stroud, Natalie J., Niche News: The Politics of News Choice, New York: Oxford
University Press, 2011.
Stroud, Natalie Jomini, and Jae Kook Lee, “Perceptions of Cable News
Credibility,” Mass Communication and Society, Vol. 16, No. 1, 2013, pp. 67–88.
Taneja, Harsh, James G. Webster, Edward C. Malthouse, and Thomas B. Ksiazek,
“Media Consumption Across Platforms: Identifying User-Defined Repertoires,”
New Media and Society, Vol. 14, No. 6, 2012, pp. 951–968.
Tsfati, Yariv, and Joseph N. Cappella, “Do People Watch What They Do Not
Trust? Exploring the Association Between News Media Skepticism and Exposure,”
Communication Research, Vol. 30, No. 5, 2003, pp. 504–529.
Twenge, Jean M., Gabrielle N. Martin, and Brian H. Spitzberg, “Trends in US
Adolescents’ Media Use, 1976–2016: The Rise of Digital Media, the Decline of TV,
and the (Near) Demise of Print,” Psychology of Popular Media Culture, August 2018.
U.S. Census, “Language Spoken at Home by Ability to Speak English for the
Population 5 Years and over,” webpage, 2017. As of August 4, 2018:
https://factfinder.census.gov/bkmk/table/1.0/en/ACS/17_1YR/B16001
Van Damme, Kristin, Cédric Courtois, Karel Verbrugge, and Lieven De Marez,
“What’s APPening to News? A Mixed-Method Audience-Centred Study on
Mobile News Consumption,” Mobile Media & Communication, Vol. 3, No. 2,
2015, pp. 196–213.
Wanta, Wayne, and Yu-Wei Hu, “The Effects of Credibility, Reliance, and
Exposure on Media Agenda-Setting: A Path Analysis Model,” Journalism
Quarterly, Vol. 71, No. 1, 1994, pp. 90–98.
Weldon, Michele, Everyman News: The Changing American Front Page, Columbia,
Mo.: University of Missouri Press, 2008.
West, Mark Douglas, “Validating a Scale for the Measurement of Credibility: A
Covariance Structure Modeling Approach,” Journalism Quarterly, Vol. 71, No. 1,
1994, pp. 159–168.

References 89
Westley, Bruce H., and Werner J. Severin, “Some Correlates of Media Credibility,”
Journalism Quarterly , Vol. 41, No. 3, 1964, pp. 325–335.
Westlund, Oscar, and Mathias A. Färdigh, “Accessing the News in an Age of
Mobile Media: Tracing Displacing and Complementary Effects of Mobile News
on Newspapers and Online News,” Mobile Media & Communication, Vol. 3,
No. 1, 2015, pp. 53–74.
Wheeless, Lawrence R., “The Effects of Attitude, Credibility, and Homophily
on Selective Exposure to Information,” Speech Monographs, Vol. 41, 1974,
pp. 329–338.
Wolf, Cornelia, and Anna Schnauber, “News Consumption in the Mobile Era,”
Digital Journalism, Vol. 3, No. 5, 2015, pp. 759–776.

www.rand.org
RR-4212-RC
9 7 8 1 9 7 7 4 0 3 4 3 8
ISBN-13 978-1-9774-0343-8
ISBN-10 1-9774-0343-3
51900
$19.00
In this report, the authors explore novel measures of how U.S. media
consumers obtain news. They examine the combinations and relative
levels of use of different news delivery platforms (e.g., print, broadcast
television, social media, internet), and the relationships between these
“news consumption profiles” and (1) consumers’ perceptions of the
reliability of news overall and of news platforms, (2) consumers’ use of
perceived reliable platforms, and (3) consumers’ willingness to seek
out news from differing viewpoints. Many people (41 percent) indicated
that they believed that news has become less reliable than in the past;
a similar number (44 percent) said they believed there has been no
change; and 15 percent said they thought news is more reliable now.
Finally, political partisanship was broadly linked to various news
consumption behaviors: consumption profiles, perceptions of news
reliability, and willingness to seek out news from differing viewpoints.

Rand rr4212 (1)

More Related Content

What's hot (19)

Similar to Rand rr4212 (1) (20)

Recently uploaded (20)

Rand rr4212 (1)