RESEARCH ARTICLE

Reliance on Facebook for news and its

influence on political engagement
Clarissa C. David ID1*, Ma. Rosel S. San Pascual1, Ma. Eliza S. Torres2
1 College of Mass Communication, University of the Philippines Diliman, Diliman, Quezon City, Philippines,
2 School of Languages, Humanities, and Social Sciences, Mapua University, Intramuros, Manila, Philippines

* ccdavid2@up.edu.ph

a1111111111 Abstract
a1111111111
a1111111111 This paper examines the link between reliance on Facebook for news, political knowledge,
a1111111111 and political engagement in the Philippines. We tested five hypotheses using data gathered
a1111111111 from an online survey of 978 Filipinos conducted from February 1 to March 31, 2016. Find-
ings support the hypothesis that those who rely less on social media as a news source
exhibit higher levels of perceived knowledge about politics than those who rely more on it for
news. Controlling for traditional news use, following political officials or institutions on social
OPEN ACCESS media is associated with higher levels of political interest and engagement, those with more
Citation: David CC, San Pascual M.RS, Torres M. politically active friends on Facebook have higher levels of exposure to political content
ES (2019) Reliance on Facebook for news and its online, and there is a positive correlation between Facebook being a source of information
influence on political engagement. PLoS ONE 14
about politics and discussing politics more often with others. However, the hypothesis that
(3): e0212263. https://doi.org/10.1371/journal.
pone.0212263 those with more friends on their network who are politically active, will have greater political
knowledge and more political engagement than those who have few politically active friends
Editor: Antonio Scala, Institute for Complex
Systems, CNR, ITALY on their Facebook network is not supported.

Received: October 8, 2017

Accepted: January 30, 2019

Published: March 19, 2019

Introduction
Copyright: © 2019 David et al. This is an open
access article distributed under the terms of the
High levels of engagement in politics are a sign of a healthy democracy, where the citizenry is
Creative Commons Attribution License, which empowered and interested in being involved in matters of the State. More than the overall
permits unrestricted use, distribution, and level of involvement in politics, it is the equitable distribution of this participation that is
reproduction in any medium, provided the original important. Inequitable political participation may result in a government that supports only
author and source are credited. the interests of those who participate. (e.g., [1])
Data Availability Statement: All relevant data are Young citizens are notoriously difficult to engage in political life, often setting aside politics
within the paper and its Supporting Information for entertainment (e.g., [2–4]) and they are often thought to be less engaged than older citi-
files. zens. In the new media environment however, young citizens have turned en masse to social
Funding: This research is supported by the Office media for news about politics and public affairs.[5]
of the Vice Chancellor for Research and Social media sites, the most popular of which is Facebook (FB), allow users to build net-
Development of the University of the Philippines, works of contacts from individual or group profiles.[6] It is a legitimate venue for engagement
Diliman through an Outright Grant to the
in civic and political life, and in the most developed countries, now an important source of
corresponding author Clarissa C. David. The
funders had no role in study design, data collection
political news. Social media’s role in the complex relationship between news use and political
and analysis, decision to publish, or preparation of engagement has become the subject of much research, with most findings suggesting that plat-
the manuscript. forms such as FB have positive influences on outcomes related to politics (e.g., [7]).

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 1 / 15

 Reliance on Facebook for news

Competing interests: The authors have declared In the Philippines, a developing country with a high poverty rate, roughly 58% are Internet
that no competing interests exist. users and more than 90% of these are on FB.[8] What are the implications of FB use for news
consumption and political participation in the context of a developing country with a tradi-
tionally low level of news use?
This paper seeks to examine the influence of reliance on FB for news on political engage-
ment, independently of use of traditional news sources. We attempt to examine various
sources of information about news and political affairs on FB, including government agency
FB pages and news sites. Specifically, it hypothesizes that for individuals who consider it a
news source, those who follow political officials and institutions on social media and those
with more politically active social media friends in their online news network are more knowl-
edgeable and more highly engaged, independently of their traditional news consumption.

News online and political engagement

There is a mature literature around the question of whether Internet use is consequential to
political engagement, and most studies conclude that its influence is generally positive. It has
been found to increase political engagement,[9–10] increase knowledge about political cam-
paigns,[11] and improve voting or other types of participation.[12–13] Internet-based expres-
sion of political views are linked to offline forms of participative behaviors.[14–18] Evidence is
likewise accumulating that these relationships extend to the social media platforms of FB and
Twitter, where users are connected as they share and engage with various kinds of content,
including news and public affairs.
Our research has a specific interest in the use of FB for news consumption, and its influence
on political engagement. The platform is now the main driver of traffic to news websites, and
while it does not generate original news content and the company has made specific state-
ments that they are not a news or media company, FB has grown into an important medium
for news.
In a 2016 study, Pew Research Center accounted that 62% of adults in the US get news from
social media.[19] Majority of the social media news consumers that they surveyed reported
that they only get their news from one social media site—FB. In fact, the same study reported
that 66% of FB users get news from FB. (ibid)
Some recent work on the information gaps created by the online news environment sug-
gests that less educated and less politically inclined citizens do not benefit with as much gain in
political knowledge from online formats compared to print, while the more educated citizens
gain knowledge regardless of the medium.[20] In the mid-2000s, scholars were already con-
cerned that the expansion of choice in news content and sources, through news websites,
would lead to a widening of knowledge gaps (e.g., [21]). Experimental and survey studies that
examine information disparities between newspaper reading and online news reading mostly
find that knowledge is better retained when reading newspapers compared to reading online
news websites. Online formats no longer had the traditional cues present in print, such as
story importance cues.[22] Moreover, organization by topical menus meant that users were
more likely to stay within their issue interests and in the end become less exposed to issues that
fall outside their core set of concerns.[23]
The entry of social media as a source of news, and the growing reliance of the general public
on these channels for news consumption, re-opens this line of inquiry. FB is considered as a
new layer of gatekeeping, its algorithm and the preferences of a user’s friends list determine
whether you see a story and how many times it will cross your feed. It remains an open ques-
tion whether political news read through FB opens up a new market for this content or only
provides an additional channel of access to those who are already news consumers in other

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 2 / 15

 Reliance on Facebook for news

media. At least for the young, those who we consider as digital natives, FB is a natural part of
their media habits and for many of them, there could be a reliance on the platform for expo-
sure to political news. It is thus as important to know the quality of their use of this media plat-
form.[24]
This study is interested in the idea of reliance on FB for information, that is, the degree
to which users are dependent on FB for news so that without it, they would have limited
exposure. We hypothesize that those who are reliant on FB were not politically engaged and
interested to begin with, and thus, would have little political knowledge outside of their FB
feed.
The empirical association between political knowledge and interest, engagement, and par-
ticipation in civic and political activities is well-established (e.g., [25–26]). Understandably,
one’s decisions concerning participation depends on relevant political knowledge.[27] The
positive relationship between Internet use and participation is stronger with interactive plat-
forms like blogs and social networks rather than static platforms online.[28] Social media use,
especially for political content, has been found to correlate with various forms of political par-
ticipation (e.g., [29–30]), although some conflicting findings also question whether these are
effects and whether they are direct or indirect.[26] Further, others contend that the Internet
has caused many to retract from public life.[12] Gil de Zuniga, Jung, and Valenzuela find evi-
dence that information seeking on social media sites are predictive of online and offline partic-
ipatory behaviors.[27] These existing studies guide our further hypothesis that following
various sources of political news on FB feeds, like government institutions, pages of politicians
and candidates, or news pages, would be associated with higher levels of political engagement
and interest, including higher levels of political discussion with others.[31–32] These associa-
tions are hypothesized to exist independently of traditional news media exposure.
Reliance on FB for political news content would mean that one’s news exposure is influ-
enced more strongly by the mechanisms on the platform that determine what kinds of news
and political content appears on a person’s news feed. The main factors that influence the con-
tent you would see on your feed may be likened to filtering: the stories that people in your
social network share, the stories that are viral or trending (popular), and those that are pushed
by paid advertising.[27] Each of these factors interact with the other, for instance news stories
that are “liked” by people in a social network and are trending in the broader environment, are
more likely to appear on your feed.
It is possible that political content exposure on social media is amplified in effect compared
to traditional media. People are able to translate their interests into participation because what
they see include opinions and emotional appeals of people that they know. We are more
inclined to pay attention to utterances from friends and family, and more likely to think more
carefully about arguments that come from them rather than strangers or even reporters.
Groups now turn to social media for recruitment, to expand their constituency.[31, 33] This
line of argument privileges social capital as a key enabler of participative behaviors, both online
and offline.[7, 27, 34]
Since content shared and posted across FB are diverse, and only a portion of it is political in
nature, we expect that those whose social networks have more politically active people will
have greater exposure to news content than others, and that this would be associated with
higher levels of political knowledge and engagement. This is consistent with findings of Wolfs-
feld, Yarchi, and Samuel-Azranal,[35] which show that social media-based political content
exposure is more highly correlated with traditional measures of political participation than
regular media. The association is much stronger even, between social media-based political
news exposure, and digital political participation.

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 3 / 15

 Reliance on Facebook for news

Facebook in the Philippines

The Philippines just entered its status as a “middle-income” country, from a long history of
being a poor country with a large proportion of the population living in poverty (around 22%).
It is a country with low levels of high school and college graduation rates and generally low
access to many forms of technology. Most recent data show laptop and desktop computer
ownership/access at only 24.3% of households across the country,[36] most schools do not
have computers in classrooms, and most students do not have access to one. However, the
rapid adoption of smartphones and the sharp decline in the costs of smartphone units that are
capable of Internet connection led to an expansion of internet access in the Philippines.
Officially, Internet penetration rate in the country is less than 40% for a country of 101 mil-
lion people. However, according to FB, there are over 50 million Filipinos in the country on
their platform. Among those that report having access to the Internet, over 94% report having
a social media account and the vast majority of them are on FB.[37] As such, the Philippines is
often characterized as the social media capital of the world, on top of its reputation as the text-
ing capital of the world. The leading telecommunication companies in the country offer access
to the FB app on smartphones without charging for data usage; this further grew the popula-
tion of Filipinos on FB.
The Philippines has a history of low levels of political knowledge acquired from traditional
media, no doubt owing to the very low circulation rates of newspapers (at best, 15% read the
paper regularly). Most news consumption happens through television, and these days, social
media. Compared to FB growth even in developed countries, FB in the Philippines grew rap-
idly into the main driver of news traffic among the most prestigious news outlets in the coun-
try. A paper (authors) on the role of FB on news reporting in the country reveals that among
the top 4 news websites, FB-referred traffic accounts for between 60%-90% of clicks into news
sites. This was a year ago and it is reasonable to think that these percentages have grown since.
Television remains the number 1 news source across the country; but among those with Inter-
net access, it is likely that the majority of news consumption is FB-driven.
This unique situation of the Philippine news reading public makes it a good venue to study
the potential implications of reliance on FB for political news. Fresh out of a contentious elec-
tion with controversial results, FB-based campaigning and political activism have flourished
and thrived, some say, to the detriment of civilized political debate. Online discussions and
comments grew vicious and sometimes violent, and full of political–and personal–vitriol, thus
recently prompting the Senate to open a hearing on the problem of social media’s effect on cul-
ture. Clearly FB has an important role to play in the Philippines’ news use, political knowledge,
and engagement with politics, providing an interesting context for understanding the relation-
ships hypothesized in this study.

Hypotheses
Based on where we are now in the literature on social media, political knowledge, and engage-
ment, this research seeks to contribute to this fast-growing area by looking specifically at FB as
a news source, reliance on it as a source of news and information, and its correlation with
political engagement.
This research is guided by an interest in answering the research question: Is FB use associ-
ated with political engagement, knowledge, and opinions about important political issues? To
systematically investigate how these concepts relate to each other, the following hypotheses are
posed:
H1: Those on social media who rely less on FB as a news source exhibit higher levels of per-
ceived knowledge about politics than those who rely more on it for news.

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 4 / 15

 Reliance on Facebook for news

Hypothesis 1 is posed in relation to a conceptualization that “reliance” means using Face-

book as a source of information without the benefit of using other traditional media sources,
such as television or newspapers. As reviewed above, the Philippine news market is not strong,
there has historically been low newspaper reading publics. Thus, high reliance individuals are
those that, in the extreme case, only get their political knowledge from Facebook while those
with low reliance may still get knowledge from the platform but also get knowledge from other
news sources. The direction of the hypothesis is consistent with extant literature showing
higher knowledge among those with higher news media consumption [10, 11].
H2: Following political officials or institutions on FB will be associated with higher levels of
political interest and engagement.
One of the ways Facebook is used by political actors and institutions, as well as media, is to
create and maintain official institutional and personal accounts which are used to publicize
programs, accomplishments, and information pertaining to an agency or a politician. We
hypothesize that when users are following such pages on FB, they are more likely to have
higher levels of engagement and interest because the active and conscious decision to click
“follow” on a page indicates a willingness to regularly see information on those pages. In cer-
tain markets where survey data are available, following politicians on social media is as high as
35% in some countries, and that the reason for following is an interest in hearing directly from
politicians and learning more detailed information [38].
H3: Having more politically active friends on FB will result in higher levels of exposure to
political content online.
Consistent with empirical literature on political discussion and online political engagement
[39], as well as what is known about the FB algorithms that prioritizes on individual newsfeeds
stories shared within networks [40], we apply these ideas to predict that having more politically
active friends on FB will result in higher exposure levels to political content. That is, if an FB
user has a large number of friends who are politically active, interested, and engaged, those
same friends will be posting and reposting about politics, news, and current affairs, which
means that even if the primary user is not necessarily politically interested, s/he would have a
higher likelihood of being exposed to political content by having online social networks that
are politically interested.
H4: Independently of traditional news consumption, those who are exposed to news
through social media will discuss politics more often with others than those who are not
exposed to news through social media.
When publics on FB encounter political information on the platform, we hypothesize that
they are more likely to engage in discussions about politics in the same space, through com-
menting and having other kinds of online and offline conversations with others about the sto-
ries they see circulating online. This hypothesis is based on what we know about exposure to
political information and political discussion, that generally, regardless of medium, there are
positive associations found in various studies which extends as well to information found
online. [41, 42]
H5: Those with more friends on their network who are politically active, will have greater
political knowledge and more political engagement than those who have few politically active
friends on their FB network.
One effect of political knowledge, discussion, and interest is political engagement, accord-
ing to historical political communication literature that examines processes of media influence
and interpersonal discussion about politics. [43, 44] Since those with politically active friends
on Facebook will see more political information as hypothesized in H4, they are hypothesized
to gain more knowledge and consequently, be more likely to engage politically.

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 5 / 15

 Reliance on Facebook for news

Method
This study is based on an online survey of Filipinos based in the Philippines. The questionnaire
contained items that measure FB use, reliance on it for news, reports of pages that they follow,
and general political interest and engagement among others. It was constructed using Survey-
Monkey which hosted the survey, data collection, and data recording. The questionnaire was
posted online through an FB page post containing the link to the questionnaire, which was
then promoted to Filipinos between the ages of 15 and 45. A Filipino version of the question-
naire and post was also promoted. Data gathering was conducted between February 1 and
March 31, 2016.
The study is an opt-in online survey with informed consent. Participants were informed
that their participation is completely voluntary, anonymous, and that they can abandon the
survey at any time even after it has been started. The University of the Philippines does not
have an Institutional Review Board, but the grant review process includes opportunity for the
granting body and the reviewers to reject the proposal based on possible harm to subjects. A
signed copy of the Waiver Certification for Ethics Review by the Vice Chancellor of Research
and Development of the University of the Philippines, whose office funded this research, is
submitted as a supporting document.
A total of 1,555 respondents started the survey; 1,439 answered in English while 116
answered in Filipino. Respondents who fell outside of the age range, those not residing in the
Philippines, and those who abandoned the questionnaire before getting to the last question,
were screened out. The final sample size is 978, where 41% are male. The average age of the
sample is 23 years and the biggest number of them completed high school (44.9%) followed by
those who completed college (33.2%). A big majority of the respondents are from Manila
(74.2%) and the rest are from other parts of the country.
Owing to the non-probability nature of the sample drawn for this study, this paper makes
no claims of representativeness to a broader population in terms of prevalence of social media
use or any other descriptive factor. The value of the following results is in the significant multi-
variate relationships tested, which still hold internal validity even when external validity is
weak. [45] The primary objective of this analysis is not to make population inferences, but to
test relationships between variables, for which, nonprobability sampling can provide data with
value particularly with exploratory research such as this one. The inability of a sample to repre-
sent a broader population “does not necessarily negate their usefulness for research”[46].

Measures
FB news reliance. Respondents were asked “If you were not able to see any news about
politics and government from FB, how informed would you be about current events?” The
response options were Not Informed at All, Only Slightly Informed, Well-Informed, and Fully
Informed.
Knowledge of politics. Self-reported political knowledge is calculated using two ques-
tions. The first is how much do respondents think they know about politics and the second
asks whether they typically turn to others for information or others turn to them for
information.
Interest in politics. Respondents answered one question on political interest on a 4-point
scale from Not at all interested in politics to Very interested in politics. The question was
“Please tell us how interested you are in national politics.”
Political engagement. Offline engagement is a standard measure of the sum of activities
respondents said they have done in the past year, including volunteering for a political organi-
zation, attending a rally, encouraging another person to vote, being an active member of a

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 6 / 15

 Reliance on Facebook for news

group that tries to infleunce public policy or government, volunteering or working for a
school-based political party, and displaying a campaign button or sticker.
Exposure to FB news. Respondents reported how many of each type of FB page they fol-
low on their FB feed, these include: government offices and agencies, political organizations,
politicians and candidates (local and national), news organizations and reporters, political
commentators, issue-based groups and other political parties or non-government organiza-
tions. For each category, the response options are 0–1, 2–4, and 5 or more. The first three cate-
gories were non-news sites; the mean was computed for these, then another mean was
computed for the news sites. The two means were summed for a variable that ranged from 2–6
(alpha = 0.85, M = 3.25, SD = 0.99).
FB Informs on politics. On a 4-point scale ranging from A Lot to Not at All, respondents
were asked if FB helps them with the following: staying informed about current events, staying
informed about the local community, learning about political issues that affect them person-
ally, learning about political issues that affect the country, and learning their friends’ political
beliefs. The scale is reliable (alpha = 0.84).
Discussing politics. Discussing politics is a mean index of three items, one that asks how
often they discuss politics with others (4-point scale from Nearly Everyday to Less Often than
a Few Times a Month), whether they lead the conversation or listen (0/1), and how much they
enjoy talking about politics with friends and family (4-point scale from Very Much to Not at
all). The final index ranges from 0.75 to 2.
Politically active friends on FB. The mean of five items was computed to represent how
many of a respondent’s friends on their FB network are politically interested and active
(alpha = 0.88). These include a question on the proportion of their FB friends who are inter-
ested in news, how many are politically active, how many frequently post or share news stories,
and how many post their personal opinions about politics.
Controls. Multivariate regressions run to test the hypotheses posed control for age, educa-
tional attainment, sex, income group, interest in politics, attention paid to politics on FB, and
how much they follow traditional news (i.e. newspaper, television, radio).

Results
In the sample, reliance on FB for news was relatively high, with 63% reporting that if they did
not obtain news from the platform they would only be slightly informed, and another 4% say-
ing they would not be informed at all. Only 33% say that without FB as a source, they would be
well informed or fully informed. The average level of self-reported political knowledge is 2.56
on a 4-point scale with higher values indicating greater knowledge (Standard deviation of the
Mean = 0.66). Twenty-eight percent (28%) say that they are usually the source of information
by others, and the rest (72%) say that they typically turn to others for political information.
Level of political interest in the sample was an average of 2.77 on a self-reported scale from 0 to
4 with higher values indicating higher interest levels (Standard deviation = 0.64). Political
engagement offline was measured by asking respondents of a list of 11 possible forms of
engagement, how many they engaged in over the past year, the possible range of the resulting
index is 0–11 and the overall mean is 3.46 with a standard deviation of 2.44.
On using FB to stay informed about politics, the respondents on average indicated they use
FB to stay informed at 2.26 (standard deviation = 0.6) on a scale ranging from 0–4 with higher
values indicating greater use of FB for political information. The computed scale for discussing
politics ranges from 0.75 to 2.75 with higher values indicating more discussion, the mean is
1.76 and standard deviation 0.5. The index computed for proportion of politically active

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 7 / 15

 Reliance on Facebook for news

friends on FB ranges from 0.2–3.8 with higher values indicating more politically interested
and active friends (M = 2.4, SD = 0.7).
Bivariate-level tests of hypotheses are reported using Pearson’s R correlation coefficients,
while multivariate tests are conducted through Ordinary Least Squares (OLS) regressions
which allows for the application of statistical controls to test robustness of bivariate associa-
tions. Table 1 reports the Pearson’s r correlations matrix for all variables in hypotheses.
Hypothesis 1 posits that those on social media who have lower reliance on FB as a news
source would exhibit higher levels of perceived knowledge about politics. Results indicate that
at the bivariate level, self-assessed knowledge of politics is moderately and positively correlated
with FB news reliance (Pearson’s r = .444� ; n = 966). The mean level of knowledge among
those who say they are fully reliant on FB for news is 4.3, while for those who would remain
well-informed even without FB as a news source is 5.69, and for those who would remain fully
informed even without FB it is 6.85. Those on FB who rely less on it as a news source exhibit
higher levels of perceived knowledge about politics than those who rely more on it for news.
Thus, at the bivariate level, those who are less reliant on FB for news perceive themselves to be
more knowledgeable about politics than those who are reliant, supporting hypothesis 1.
To examine the robustness of the association, it was tested in an ordinary least squares
regression where knowledge is the main dependent variable and reliance is the main indepen-
dent variable, other variables were added as controls or as part of the hypothesized effects
(Table 2). OLS Regression was run predicting self-assessed political knowledge (Table 2) with
news reliance, FB use measures, news exposure measures, and other relevant variables and
controls. Each of the three models adds new variables as controls. Model 1 controls for FB-
related variables, Model 2 adds controls for following traditional news, interest in poltiics, and
discussing politics, while Model 3 adds controls for all demographic variables. Adding each set
of controls in different models provides additional information in terms of which controls
were likely reducing the predictive value of the main independent variable.
OLS regressions generate two coefficients, Model “Bs” and model betas, the latter is a stan-
dardized version of the former. Model “Bs” are changes in the independent variable resulting
from changes in the dependent variable, expressed in the unit of the dependent variable. That
is, for each unit increase in FB news reliance, there is a 0.61 unit increase in perceived political
knowledge, controlling for all other variables in the model. Model betas are standardized par-
tial correlation coefficients between the independent variable and the dependent variable, con-
trolling for the other variables in the model.

Table 1. Pearson’s r correlation coefficients.

(1) (2) (3) (4) (5) (6) (7) (8) (9)
(1) Reliance on FB 1
(2) Perceived knowledge .444��� 1
�� ��
(3) Following news sites and govt agencies on FB .291 .389 1
(4) Interest in politics .376�� .415�� .464�� 1
(5) Offline participation (Engagement) .219�� .327�� .483�� .433�� 1
(6) Traditional news use .394�� .361�� .381�� .411�� .251�� 1
(7) Having politically active friends on FB .043 .272�� .285�� .270�� .226�� .191�� 1
(8) FB is source of information .062 .312�� .305�� .353�� .212�� .218�� .352�� 1
(9) Discussing politics with others .396� .532�� .455�� .551�� .434�� .361�� .272�� .312�� 1
�
p < .05
��
p < .01
���
p < .0001

https://doi.org/10.1371/journal.pone.0212263.t001

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 8 / 15

 Reliance on Facebook for news

Table 2. OLS Regression coefficients predicting self-assessed political knowledge.

Model 1 Model 2 Model 3
B Beta B Beta B Beta
2.487 2.114 1.772
FB News Reliance 0.601 0.28� 0.469 0.218� 0.386 0.179�
� �
Heavy FB user 0.148 0.076 0.127 0.065 0.083 0.043
Uses FB Often -0.007 -0.091� -0.005 -0.071� -0.004 -0.056
Follows news sites and gov’t agencies on FB 0.244 0.189� 0.174 0.135� 0.21 0.163�
Political engagement on FB 0.075 0.026 -0.013 -0.005 -0.027 -0.009
Attention to politics on FB 0.117 0.126� 0.012 0.013 0.01 0.011
FB Informs on politics 0.063 0.03 0.014 0.007 0.011 0.005
FB friends with different views 0.103 0.047 0.1 0.046 0.073 0.034
FB friends politically active -0.011 -0.006 -0.046 -0.025 -0.027 -0.015
Follows traditional news 0.043 0.024 -0.026 -0.014
Interest in politics 0.042 0.02 0.026 0.013
Discusses politics 0.76 0.302� 0.795 0.315�
Age 0.031 0.159�
Education 0.014 0.014
Female -0.274 -0.106�
Income group 0.059 0.04
R2 0.24 0.298 0.33
Adjusted R2 0.232 0.298 0.318
�
p < .05
��
p < .01
���
p < .0001

https://doi.org/10.1371/journal.pone.0212263.t002

With all controls applied, there is a remaining significant association between reliance on
FB for news and self-assessed political knowledge (beta = 0.179� ). In the same model, the
strongest associate of knowledge is discussing politics (beta = 0.315� ) followed by following
news and government agency pages on FB (beta = 0.163� ). Among demographic controls,
males and younger respondents are less likely to be politically knowledgeable. The total model
explains 33% of the variance of self-assessed political knowledge.
Hypothesis 2 posits that following political officials or institutions on social media will be
associated with higher levels of political interest and engagement. Respondents that report fol-
lowing more FB pages of government agencies and news websites are more likely to say they
are interested in politics (Pearson’s r = 0.464; n = 925), more likely to engage in offline political
activities (Pearson’s r = 0.3434; n = 823), and more likely to display online political engage-
ment (r = 0.483; n = 886). Controlling for following news on traditional channels and other
variables in the model, hypothesis 2 is supported.
With controls for all the same variables as in Table 2, the relationship between following
news pages online and political engagement remains significant such that those who report fol-
lowing government agency and news sites on FB tend to be more politically engaged offline
(b = 0.267� ). The full model results can be found in Table 3, wherein the column for Model 4
coefficients reflects results controlling for demographic variables.
Hypothesis 3 is likewise supported such that controlling for traditional news use, those with
more politically active friends on FB have higher levels of exposure to political content online,
as measured by whether FB informs them about politics (Pearson’s r = 0.352; n = 947).

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 9 / 15

 Reliance on Facebook for news

Table 3. OLS Regression coefficients predicting political engagement offline.

Model 1 Model 2 Model 3 Model 4
B Beta B Beta B Beta B Beta
-1.8 -2.917 -3.124 -3.83
FB News Reliance 0.305 0.076� 0.103 0.026 0.052 0.013 -0.018 -0.005
Heavy FB user 0.176 0.046 0.135 0.035 0.118 0.031 0.086 0.023
� �
Uses FB Often 0.009 0.059 0.011 0.076 0.011 0.078 0.012 0.083�
� � �
Follows news sites and gov’t agencies on FB 0.761 0.308 0.643 0.26 0.625 0.253 0.66 0.267�
Political engagement on FB 0.235 0.043 0.207 0.038 0.206 0.037 0.239 0.043
Attention to politics on FB 0.199 0.109� 0.022 0.012 0.021 0.012 0.017 0.009
FB Informs on politics 0.073 0.018 -0.023 -0.006 -0.024 -0.006 -0.058 -0.014
FB friends with different views -0.003 -0.001 -0.013 -0.003 -0.029 -0.007 -0.052 -0.012
FB friends politically active 0.099 0.027 0.079 0.022 0.088 0.024 0.065 0.018
Follows traditional news -0.078 -0.022 -0.083 -0.024 -0.115 -0.033
Interest in politics 0.654 0.159� 0.648 0.157� 0.626 0.152�
Discussing politics 0.663 0.134� 0.584 0.118� 0.635 0.128�
Political knowledge 0.105 0.055 0.08 0.042
Age 0.017 0.047
Education 0.092 0.051
Female 0.06 0.012
Income group 0.014 0.005
R2 0.228 0.261 0.263 0.27
Adjusted R2 0.219 0.25 0.251 0.255
�
p < .05
��
p < .01
���
p < .0001

https://doi.org/10.1371/journal.pone.0212263.t003

The same holds true at the bivariate level with Hypothesis 4, whereby there is a positive cor-
relation between FB being a source of information about politics, and discussing politics more
often with others (r = 0.312, n = 947). Table 4 shows regression results for models predicting
political discussion. It is evident that with multiple controls, there is no longer a significant
association between FB being a source of information about politics, and political discussion.
Significant predictors of discussing politics with others are non-reliance on FB for news, pay-
ing attention to politics on FB, being interested in politics, and following news site pages on
the social media platform.
Hypothesis 5 posits that those with more politically active friends on their network will
have greater political knowledge and more political engagement than those who have few
politically active friends on their FB network. Tables 2 and 4 show that this hypothesis is not
supported. There is no significant relationship between having a larger number of politically
active friends on FB and being more politically knowledgeable and engaged.

Discussion
In this research we seek to examine how “reliance on FB” for political news influences citizens’
political knowledge and participation. We understand reliance to mean that if one does not
receive news stories through the FB platform, there would be no alternative source and the
result is low knowledge of politics. There was a special interest in capturing these associations
among relatively young people, those below the age of 45 years old.

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 10 / 15

 Reliance on Facebook for news

Table 4. OLS Regression coefficients predicting political discussion.

Model 1 Model 2 Model 3
B Beta B Beta B Beta
(Constant) 0.406 0.189 0.23
FB News Reliance 0.144 0.169� 0.097 0.114� 0.104 0.122�
Heavy FB user 0.024 0.031 0.019 0.024 0.023 0.03
Uses FB Often -0.002 -0.055 -0.001 -0.037 -0.001 -0.04
Follows news sites and govt agencies on FB 0.079 0.154� 0.05 0.097� 0.043 0.085�
� �
Political engagement on FB 0.106 0.092 0.11 0.095 0.11 0.095�
� �
Attention to politics on FB 0.128 0.348 0.098 0.266 0.098 0.267�
FB Informs on politics 0.05 0.061 0.022 0.027 0.023 0.028
FB friends with different views 0.003 0.004 0 0.001 0.002 0.002
FB friends politically active 0.042 0.057 0.035 0.048 0.031 0.043
Follows traditional news 0.024 0.033 0.033 0.045
Interest in politics 0.191 0.233� 0.191 0.234�
Age -0.005 -0.071�
Education 0.013 0.035
Female 0.014 0.014
Income group -0.006 -0.01
R2 0.394 0.428 0.431
Adjusted R2 0.387 0.421 0.421
�
p < .05
��
p < .01
���
p < .0001

https://doi.org/10.1371/journal.pone.0212263.t004

The online survey of young Filipinos on FB result in support for most hypotheses posed in
this study. Respondents who say they are fully reliant on the platform for news, think of them-
selves as less politically knowledgeable than those who say they are not reliant. This association
holds even while controlling for news consumption in traditional news channels. Following
the FB pages of political officials or institutions on social media is associated with higher levels
of interest and engagement among respondents. Moreover, respondents who report having
more friends who are politically active report higher levels of exposure to political content
online.
The hypothesis that those who see themselves as sources of political information during
conversations are also more likely to discuss politics with others was not supported by the
data. Also not supported was the hypothesis that those with more friends on their network
who are politically active will have higher levels of knowledge and engagement than those who
have fewer politically active friends on FB.
FB’s growth as a significant player in the news market does not happen in a vacuum without
the convergence of certain factors. It happens within the context of a continued existence of
traditional news, news consumer’s social networks, and an effort among news sources (e.g.
government agencies) to engage directly with their constituents through the platform. This
research attempts to capture some portion of this dynamic.
While reliance on FB for news is positively associated with self-assessed political knowledge,
it is not a significant predictor of engagement. People do receive information from FB, and for
some of them it is their only source of news. Without FB, it is possible that these individuals
would have no other source of news and thus, would have lower levels of political knowledge.

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 11 / 15

 Reliance on Facebook for news

In that sense, FB is bringing in new audiences for public affairs, through more incidental
forms of exposure in FB feeds than through information-seeking behaviors such as checking
news websites. Those who follow official government pages or official news pages on FB are
more knowledgeable, even when controlling for reliance and other measures in this study.
Thus, the remaining significant effect of reliance on knowledge captures incidental exposure,
the kind that is not a result of stories appearing across one’s news feeds after having followed a
news outlet’s FB page. If this interpretation is correct, then it is theoretically consistent with
the lack of influence of FB reliance on political engagement. The reliant are those who have lit-
tle interest to begin with, they see stories on their feed only because their social network share
them (in the extreme scenario), and thus, they are least likely to have interest in political
action.
The lack of support for hypothesized associations between political activity of friends on FB
and political knowledge and engagement is curious. Momentarily setting aside the possibility
that the reason is methodological, the apparent lack of significant relationship is intriguing. If
you have many friends on your feed who are politically active and interested, it stands to rea-
son that either through persuasion or exposure or both, you would be more willing to engage
in political action. The absence of a correlation, where the significant correlates to offline polit-
ical engagement are interest and following news, suggests that FB may not be an effective way
to mobilize those whose current interest in politics is either null or weak.
There are a number of limitations to this study, foremost of which is its reliance on self-
reports on knowledge and reliance on FB for news. Studies that have examined different mea-
sures of political knowledge [47] conclude that “knowledge tests” are the best most objective
indicators, and self-reported ones are questionable in their ability to capture real knowledge.
That said, the results reported in Table 1 suggests that the measure captures a substantial
amount of variance, such that it correlates with variables that it is usually associated with (e.g.
discussing politics, age, sex). In other words, while not the ideal measure, it captures political
knowledge in some way. Follow-on research could focus on greater methodological work in
measuring “reliance” using a more expansive battery of questions that capture all other poten-
tial sources of political information.
Since the study’s respondents are only Filipino audiences, the degree to which findings are
applicable to outside markets remains an open question. However, we submit that the reliance
on Facebook for news delivery and the influence it has had on the nature of the news consum-
ing public is felt in many countries in the world. Replications of the same hypothesis tests,
using the same questionnaire, in different markets are welcome and would provide a nuanced
cross-country picture of the scope and breadth of the impact that FB has in individual
countries.
This research presents possible take-off points, especially as FB itself makes stronger its
claim of being the main portal through which citizens learn about public affairs, discuss it,
organize action around it, and define themselves as citizens. Various lines of inquiry on the
implications of FB on news production and consumption remain open, with the former com-
manding greater attention and faster development in the literature. The evolution of news con-
sumption through FB and its effect on the political opinions and behaviors of citizens varies
across countries of differing political regimes, levels of development, and access to technology.
In a country like the Philippines where the population does not have a strong history of
journalism and news use, FB’s role in the shaping of politics and citizenship is likely going to
be bigger, and will happen faster than in more developed and stable democracies. The potential
of FB to bring in and create larger audiences for news is strong, but this has also meant that the
bar for what counts as “news” is low for most of these new audiences. It is an evolution of

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 12 / 15

 Reliance on Facebook for news

citizenship and political engagement that will be influenced disproportionately by FB, what it
decides to do, and how it decides to define its role in the news media landscape.

Supporting information
S1 File. Survey questionnaire.
(PDF)
S2 File. Survey dataset.
(CSV)

Author Contributions
Conceptualization: Clarissa C. David.
Formal analysis: Clarissa C. David, Ma. Rosel S. San Pascual.
Funding acquisition: Clarissa C. David.
Investigation: Ma. Rosel S. San Pascual, Ma. Eliza S. Torres.
Methodology: Clarissa C. David.
Project administration: Ma. Eliza S. Torres.
Writing – original draft: Clarissa C. David.
Writing – review & editing: Clarissa C. David.

References
1. Burns N, Schlozman KL, Verba S. The private roots of public action: gender, equality, and political par-
ticipation. Cambridge, MA: Harvard University Press; 2001.
2. Bennett WL, Xenos M. Young Voters and the Web of Politics: Pathways to Participation in the Youth
Engagement and Electoral Campaign Web Spheres, Working Paper # 20 [Internet]. Center for Informa-
tion and Research on Civic Learning and Engagement [cited 26 Oct 2017]; 2004. Available from: http://
www.civicyouth.org/PopUps/WorkingPapers/WP20Bennett.pdf
3. Bennett WL, editor. Civic life online: Learning how digital media can engage youth [Internet]. Cam-
bridge: MIT Press; 2008. Available from: https://mitpress.mit.edu/sites/default/files/titles/free_
download/9780262524827_Civic_Life_Online.pdf
4. Levine P, Lopez MH. Youth Voter Turnout has Declined by any Measure [Internet]. Center for Informa-
tion and Research on Civic Learning and Engagement; 2002. Available from: http://pweb.jps.net/~
lsbonnin/mark/docs/MeasuringYouthVoterTurnoutFinal.pdf
5. Mitchell A, Gottfried J, Matsa KE. Millennials and Political News [Internet]. Pew Research Center’s
Journalism Project. 2015 [cited 2017Sep26]. Available from: http://www.journalism.org/2015/06/01/
millennials-political-news/
6. Boyd D. Why youth (Heart) social network sites: The role of networked publics in teenage social life. In
Bukingham D, editor. Youth, Identity, and Digital Media. Cambridge, MA: MIT Press; 2008.
7. Valenzuela SCA, Park N, Kee KF. Is There Social Capital in a Social Network Site?: Facebook Use and
College Students Life Satisfaction, Trust, and Participation. Journal of Computer-Mediated Communi-
cation. 2009; 14(4):875–901.
8. Digital 2017 Global Overview Report. https://www.rappler.com/technology/features/159720-ph-
spends-most-time-online-and-on-social-media-report
9. Johnson TJ, Kaye BK. Around the World Wide Web in 80 Ways. Social Science Computer Review.
2003; 21(3):304–25.
10. Kenski K, Stroud NJ. Connections Between Internet Use and Political Efficacy, Knowledge, and Partici-
pation. Journal of Broadcasting & Electronic Media. 2006; 50(2):173–92.
11. Drew D, Weaver DH. Voter learning and interest in the 2004 Presidential election: Did the media mat-
ter? Journalism and Mass Communication Quarterly. 2006; 83(1):25–53.

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 13 / 15

 Reliance on Facebook for news

12. Quintelier E, Vissers S. The Effect of Internet Use on Political Participation. Social Science Computer
Review. 2008; 26(4):411–27.
13. Tolbert CJ, Mcneal RS. Unraveling the Effects of the Internet on Political Participation? Political
Research Quarterly. 2003; 56(2):175.
14. Zúñiga HGD, Puig-I-Abril E, Rojas H. Weblogs, traditional sources online and political participation: an
assessment of how the internet is changing the political environment. New Media & Society. 2009; 11
(4):553–74.
15. Gustafsson N. New hopes for democracy or a pirated elite? Swedish social media users and political
mobilization. Paper presented at the Western Political Science Association Annual Conference; 2010
April; San Francisco.
16. Nah S, Veenstra AS, Shah DV. The Internet and anti-war activism: A case study of information, expres-
sion, and action. Journal of Computer-Mediated Communication. 2006; 12(1):230–247.
17. Puig-i-Abril E, Rojas H. Internet use as an antecedent of expressive political participation among early
Internet adopters in Colombia. International Journal of Internet Science. 2007; 2,28–44.
18. Rojas H, Puig-I-Abril E. Mobilizers Mobilized: Information, Expression, Mobilization and Participation in
the Digital Age. Journal of Computer-Mediated Communication. 2009; 14(4):902–27.
19. Gottfried J, Shearer E. News Use Across Social Media Platforms 2016 [Internet]. Pew Research Cen-
ter’s Journalism Project. 2016 [cited 2017Oct26]. Available from: http://www.journalism.org/2016/05/
26/news-use-across-social-media-platforms-2016/
20. Yang J, Grabe ME. Knowledge acquisition gaps: A comparison of print versus online news sources.
New Media & Society. 2011Apr; 13(8):1211–27
21. Jerit J, Barabas J, Bolsen T. Citizens, knowledge, and the information environment. American Journal
of Political Science. 2006; 50(2):266–282.
22. Althaus SL, Tewksbury D. Agenda Setting and the “New” News. Communication Research. 2002; 29
(2):180–207.
23. Eveland WP, Marton K, Seo M. Moving beyond “Just the Facts”. Communication Research. 2004; 31
(1):82–108.
24. Brandtzaeg PB. Social Networking Sites: Their Users and Social Implications—A Longitudinal Study.
Journal of Computer-Mediated Communication. 2012; 17(4):467–88.
25. Bimber B, Copeland L. Digital Media and Political Participation Over Time in the US: Contingency and
Ubiquity. Paper presented at the Annual Meeting of the European Consortium for Political Research;
2011 August 4; Reykjavik, Iceland. Available from: https://ecpr.eu/Filestore/PaperProposal/db466b6c-
9c16-4491-b92a-32f1b0e80273.pdf
26. Boulianne S. Online news, civic awareness, and engagement in civic and political life. New Media &
Society. 2016Oct; 18(9):1840–56.
27. Zúñiga HGD, Jung N, Valenzuela S. Social Media Use for News and Individuals Social Capital, Civic
Engagement and Political Participation. Journal of Computer-Mediated Communication. 2012; 17
(3):319–36.
28. Flew T. New Media: an Introduction. Melbourne: Oxford University Press; 2008.
29. Dimitrova DV, Shehata A, Strömbäck J, Nord LW. The Effects of Digital Media on Political Knowledge
and Participation in Election Campaigns. Communication Research. 2011Feb; 41(1):95–118.
30. Skoric MM, Zhu Q, Goh D, Pang N. Social media and citizen engagement: A meta-analytic review. New
Media & Society. 2016Sep; 18(9):1817–39.
31. Bennett WL. The Personalization of Politics. The ANNALS of the American Academy of Political and
Social Science. 2012Mar; 644(1):20–39.
32. Bimber B. Digital media and citizenship. In: Semetko HA, Scammell M, editors. The SAGE handbook of
political communication. Los Angeles: SAGE; 2012. 115–127.
33. Bimber BA, Flanagin AJ, Stohl C. Collective action in organizations: interaction and engagement in an
era of technological change. New York: Cambridge University Press; 2012.
34. Putnam RD. Bowling alone: the collapse and revival of American community. New York, NY: Simon &
Schuster; 2007.
35. Wolfsfeld G, Yarchi M, Samuel-Azran T. Political information repertoires and political participation. New
Media and Society. 2015;1–20
36. Official Website of International Telecommunications Union, accessed 26 October 2016
37. Center for Media Freedom and Responsibility. Know your source: Think before you click; 2016 Oct 4
[Updated 2017 Jan 3; Cited 27 October 2016]. Available from: http://cmfr-phil.org/in-context/knowing-
your-source-think-before-you-click/

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 14 / 15

 Reliance on Facebook for news

38. Karogelopolous A. (2017). Following Politicians in Social Media. Reuters Institute. http://www.
digitalnewsreport.org/survey/2017/following-politicians-social-media-2017/
39. Zhang W., Johnson T.J., Seltzer T., & Bichard S.L. (2010). The revolution will be networked: The influ-
ence of social networking sites on political attitudes and behavior. Social Science Computer Review,
28(1):75–92. https://doi.org/10.1177/0894439309335162
40. DeVito M.A. (2016). From Editors to Algorithms: A values-based approach to understanding story
selection in the Facebook news feed. Digital Journalism, 5(6), 753–773. https://doi.org/10.1080/
21670811.2016.1178592
41. de Zuniga H.G., Jung N. & Valenzuela S. (2012). Social media use for news and individuals’ social capi-
tal, civic engagement, and political participation. Journal of Computer-Mediated Communication,
https://doi.org/10.1111/j.1083-6101.2012.01574.x
42. Bakker T.P. & de Vreese C.H. (2011). Good news for the future? Young people, internet use, and politi-
cal participation. Communication Research, 38(4):451–470. https://doi.org/10.1177/
0093650210381738
43. Kwak N., Williams A.E. & Lee X.W.H. (2005). Talking politics and engaging politics: An examination of
the interactive relationships between structural features of political talk and discussion engagement.
Communication Research, (32) 1: 87–111. https://doi.org/10.1177/0093650204271400
44. Xenos M. & Moy P. (2007). Direct and differential effects of the Internet on political and civic engage-
ment. Journal of Communication, https://doi.org/10.1111/j.1460-2466.2007.00364.x
45. Baker, R., Brick, J.M., Bates, N.A., Battaglia, M., Couper, M.P., Dever, J.A., Gile, K.J., & Tourangeu, R.
(2013) Non-probability sampling: Report of the Americal Association for Public Opinion Research on
non-probability sampling. Retrieved on December 1, 2018 from https://www.aapor.org/Education-
Resources/Reports/Non-Probability-Sampling.aspx#FIT%20FOR%20PURPOSE.
46. Fricker, R.D. Sage Handbook of Online Research Methods. Chapter 11. Sampling Methods for Web
and E-Mail Surveys, Chapter 11. (p.205, retrieved on December 8 from https://study.sagepub.com/
sites/default/files/Fricker.pdf)
47. Delli Carpini M.X. and Keeter S. (1996). What Americans know about politics and why it matters. Yale
University Press, NY.

PLOS ONE | https://doi.org/10.1371/journal.pone.0212263 March 19, 2019 15 / 15