Varying amounts of information in health news headlines can affect user selection and interactivity

Abstract

This experimental study (N = 308) examined whether informativeness of headlines affects user engagement and interactivity. Engagement was measured by offering participants 32 pairs of headlines manipulated with the amount of information and the reasons why the preferred headline was selected. To measure interactivity, participants were asked to indicate whether they would likely share, like or comment on the news represented by the selected headline. Results indicated a statistically significant preference for more informative headlines versus “teasing” headlines, that the level of a headline’s selection was the primary reason for selection and that Varying amounts of information in health news headlines can affect user selection and interactivity.

Full text

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
Varying amounts of information in health news headlines
can affect user selection and interactivity

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
1
INTRODUCTION
A 2017 Pew survey measuring news consumption in the U.S. found about 70% of
Americans were interested in news about health and medicine. This level of interest, however,
does not always translate into higher levels of news consumption overall. That is because on
average, only 17% of users actively seek health news (Funk, Gottfried, & Mitchell, 2017). This
gap suggests that while news consumers may express an interest in health news, fewer actively
search for health news. Arguably, users exposed to health news online may or may not engage
with it depending on how many headlines are presented.
Dor (2003) suggested that headlines play the role of a “textual negotiator” between a
news story and its readers because every headline invites readers to select and read specific
stories. This negotiator role for headlines offers new reasons to empirically explore whether
certain headline characteristics might entice more readers to select and read the related health
stories. In that context, this study proposes that a headline is a vantage point to engage more
readers with more health news.
Specifically, this study examines the effects of varying amounts of key information
presented in health news headlines. News media vying for readers’ attention in digital spaces
have increasing competition from countless sources. Consequently, many organizations are
applying various production techniques with the hope to engage more readers. These techniques
include varying the information presented in headlines. Previous research found that some U.S.
news media have been producing more “explanatory headlines” (Boland, 2017). Applying on
that strategy, the goal of this experimental study is to investigate if varying amounts of headline
information could be the primary reason that significantly more users select corresponding health
stories.

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
2
Heath information is the focus because health news is an increasingly important yet often
complex news domain that affects individuals of all demographics. Another reason for focusing
on health news is that some consumers are exposed to confusing or misleading health news
online. Previous research found that misleading health-related information can garner more
attention than evidence-based reports (Forster, 2017; Raphael, 2019). This can be more
problematic when the general public encounters news online and takes action without confirming
the source of the information (Shapiro, 2018). In this context, it would be beneficial to increase
the public’s engagement with accurate and informative health news.
Informativeness in headlines
Traditionally, the field of journalism often defined the formula for important information
as “the five Ws” or who, what, when, where and why plus how. Accordingly, this study employs
the five Ws formula to define the level of headline “informativeness.” Specifically, headline
informativeness is represented by two categories. Highly informative headlines include three or
more of the five Ws plus how. Conversely, low informative headlines contain no more than two
of the five Ws plus how. Based on that explication, this study asks,
RQ1: Does the level of information in health news headlines
significantly affect user selection?
Headline research by Dor (2003), Geer and Kahn (1993), and Ifantidou (2009) examined
the role of headlines but did not test the effects of varying amounts of information. Other
research, such as Growney and Hess (2019) found that positive and negative health news can
affect the moods of younger and older adults. Lopez, Prince, and Roche (2014) also examined
the uses of nouns in headlines to provide cues of informativeness. The researchers proposed

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
3
guidelines to generate enticing headlines without the loss of informativeness. Although these
studies addressed informativeness in news articles, their designs differ from the present study.
One difference is that the theoretical foundation of this study posits that user attention to
many headlines is often the result of incidental exposure. Theoretically, scholars explained
incidental exposure to online content such as headlines occur without a user’s prior intention
(Tewksbury, Weaver, & Maddex, 2001; Lee, 2009; Nguyen, 2008).
CONCEPTUALIZATION
Incidental exposure
Since the early research of incidental exposure by Downs (1957), the theory has become
more applicable to online journalism. Downs conceptualized incidental exposure as a by-product
of individuals’ non-political activities and more recent studies argued that incidental exposure to
online content does not cost other efforts to seek and find information (Yadamsuren & Erdelez,
2010). The notion of incidental exposure suggests that there is a potential for an online user to
encounter unexpected content as they browse or scan content (Tewksbury, Weaver, & Maddex,
2001).
Prior to the advent of the internet, journalism scholars often applied incidental exposure
to content in print and broadcast media. Specifically, Zukin and Snyder (1984) researched the
passive learning by news consumers from information in newspapers and on television. Baum
(2002) examined the incidental exposure to foreign affairs content on television.
The rapid rise of users on the World Wide Web in the mid-1990s, and later on social
media, has prompted the efforts for some scholars to investigate the effects of incidental
exposure to digital information. Since social media also make it easier for users to be exposed to

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
4
varied online news, researchers have examined user interest related to the incidental exposure to
news on social media (Boczkowski et al., 2017; Fletcher and Nielsen, 2018; Valeriani and
Vaccari, 2016). Others examined the interplay between incidental exposure to news with user
variables such as age, gender, and technology used (Tewksbury et al., 2001, Lee, 2009; Mitchell
et al., 2017).
Social media can also produce incidental exposure to news as users share and discuss
news content with others. This sharing includes links to digital-only news media, which are often
displayed as just “previews” with only headlines, text snippets and multimedia. Shared content is
instantly and incidentally exposed to recipients even if they do not click a related link for more
information (Fletcher & Nielsen, 2018).
Other scholars examined the benefits to users incidentally exposed to news. Yadamsuren
and Erdelez (2010) argued that incidental exposure to news has become a way for many users to
be informed about current events. The study also found a significant positive correlation between
incidental exposure to news and user engagement with that information.
Fletcher and Nielsen (2018) noted that exposure to news on social media is a by-product
of the constant use of social media on smartphones. The study also found that regular social
media users incidentally exposed to news depend on significantly more sources for information
than those who do not regularly use social media (Fletcher & Nielsen, 2018). In summary, there
is ample evidence of the effects of incidental exposure to digital information. This study seeks to
expand the literature by focusing on the effects of incidental exposure to health news and
whether there are primary reasons why headlines with either high or low informativeness are – or
are not - selected. The second research question is:

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
5
RQ2: Are there primary reasons for users’ selection of informative
versus less informative health news headlines?
In addition to identifying any significant reasons why certain headlines are selected, this
study also explores possible relationships between user selection and interactivity (i.e. sharing,
liking, commenting) with headlines that vary with the amount of information.
Interactivity
The importance of user interactivity with online news content continues to increase. Prior
to the digital technology that instantly connects us, traditional audiences were considered more
passive, or simply receiving one-way messages from a limited number of newspapers, radio and
television stations. This process was often conceptualized as “the magic bullet” or “inoculation”
of media. The debate about passive versus active audiences evolved since the 1950s, however,
when the early media researchers Elihu Katz, Jay Blumer, and Denis McQuail challenged the
presumed power and the effects of mass media by proposing new research recognizing that news
consumers actively make their own choices. This approach was built upon Herta Herzog’s
(1941) early paradigm termed “uses and gratifications.”
In a contemporary context, Williams (2003) states that uses and gratifications is based on
three basic assumptions: 1) people are active users when they use media, 2) people are aware of
their motivation of using the media and can explain their motivations, and 3) common patterns of
media consumption and usage can be identified among users.
The concept has been expanded to social media. Napoli (2011) suggested that
engagement is a broader phenomenon that describes various types of user attention and
involvement with content. The concept of engagement now includes the involvement, real or
perceived, of users producing, consuming, or disseminating information. This involvement is

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
6
often referred to as interactivity, which serves as a fundamental dimension of the concept of
engagement (Ksiazek, Peer, & Lessard, 2016).
Research on interactivity has highlighted the multi-directional flow of information. The
interactive nature of online media enables the audience to not only receive information but also
revise and disseminate it (Jenkins, 2006; Thurman, 2008). It is commonly known that the
interactive features of online media enable audiences to like, comment on and share content with
other users, sometimes a mass audience. Audience analytics offer news organizations the
opportunity to directly measure the audience’s engagement with - and the effectiveness of - their
content. Many newsrooms also embrace readers’ comments with the goal to improve the quality
of their news. Accordingly, this study seeks to understand if the level of interactivity is related to
headlines that vary in levels of informativeness.
RQ3: Does user interactivity (sharing, liking, commenting) with more
informative health news headlines differ from interactivity with less
informative headlines?
METHOD
Design
This study employed an online experiment to test whether levels of headline
informativeness affect user selection and why. The study also tests if headlines varying in levels
of informativeness result in more or less sharing, liking or commenting.
Measures
Dependent variables included participants’ headline selection, reason(s) for selecting a
headline, and whether participants would share, like or comment on content related to the
selected headline.

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
7
Stimuli and Pre-testing
The original headlines tested were randomly selected from six mainstream news sources
including: two networks (CNN and Fox), two legacy print sources (The New York Times and the
Washington Post) and two popular but non-traditional sources (BuzzFeed and The Huffington
Post). A total of 120 headlines collected (20 from each news organization) were coded for levels
of informativeness. Two trained independent coders used an online form to record the number of
the five “Ws” (who, what, where, when and why) plus “how” supplied in each headline. Initial
intercoder reliability (Scotts Pi = .88) reached the threshold of acceptance for reliability but
further discussion between coders following the first 10% of the headlines coded improved the
reliability (Scotts Pi = .99). Headlines for the same story were then paired with one headline of
low informativeness and one headline of high informativeness.
To confirm the informativeness in the headlines, a pre-test was conducted with an
independent sample of eight graduate journalism students who indicated which headlines in each
pair were most informative. All but three pairs of the headlines test were confirmed to differ in
levels of informativeness as coded. Three pairs of headlines deemed as “no difference” were
rewritten based on the feedback from pre-test participants.
Instrument
Before viewing the headlines in the online experiment, a five point Likert scale was used
to measure users’ level of interest in five news categories including: entertainment, sports,
health, science, and politics. The online survey included eight headline pairs (from eight news
stories) where each of the pairs has a low informative headline and a high informative headline.
Two questions followed each headline pair. The first asked participants to indicate if primary
reason for selecting the headline was more informative, interesting or personalized or raised

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
8
curiosity or for another reason. The second question asked whether participants were most likely
to share, like, and/or comment about content related to the headline. Finally, demographic
questions of age, gender and level of completed education completed the instrument.
Procedure
An email inviting participants 18 years and older was distributed to students in
journalism and information science at a large Mid Atlantic University, offering the opportunity
to be randomly selected for one of two $25 Amazon gift cards. Following the website’s
“welcome” page, users wishing to participate provided an IRB-approved consent then proceeded
to the instructions informing participants they would be exposed to several pairs of news
headlines. Participants were asked to read the headlines as they would typically read headlines
online then answer two questions after each pair of headlines.
To enhance validity of the online study, participants were prohibited from completing the
experiment more than once (per IP address) plus the “back” or “save” buttons were removed to
prohibit participants from changing previous answers. Upon completion a “thank you” page
invited participants to share the survey link with family and friends via email or social media.
RESULTS
The sample (N = 308) included 118 men (41%) and 155 women (54%). Thirty-five did
not indicate their gender. Mean age was 30.3 (SD = 11.1) ranging from 18 to 66 years. Average
time to complete the survey was 10.4 minutes. For education completed, 52.3% indicated a
bachelor’s degree, 19.2% master’s, 18.8% a high school diploma, 7% associate’s degree and
2.8% doctorate or JD. Consuming news online, most (47.4%) indicated “often” followed by
33.7% “occasionally”, and 11.6% “hourly.” Nearly 75 participants said they rarely consumed
news online.

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
9
Headline informativeness
RQ1 asked whether the level of information in health news headlines significantly affect
user selection. Table 1 displays the percentage of times that either a highly informative headline
or a low informative headline was selected when presented in pairs. For pairs presented from
eight health news stories, the percentage of participants selecting highly informative headlines
was consistently higher than the selection of low information headlines. The overall mean
selecting high information headlines (n = 70.9) was 63.7% compared to the mean percentage of
low information headlines of 36.3% (n = 45.5). A t test indicated the difference to be statistically
significant (t = 6.2, p = <.001).
Table 1.
Percentage of selection of high information versus low information headlines
Article ID
High informative
Low informative
n
%
n
%
1
97
69.8
42
30.2
2
64
58.7
45
41.3
3
80
59.3
55
40.7
4
71
55.0
58
45.0
5
74
68.5
34
31.5
6
88
69.8
38
30.2
7
100
76.9
30
23.1
8
65
51.18
62
48.82
M
79.9
63.7
45.5
36.3

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
10
Reason for selection
Comparing the reasons for selecting the preferred headline (Table 2), the majority (n =
449) of participants (48.0%) indicated because it was “more informative.” 21.1 percent (n = 197
chose a headline because it “raised curiosity” and 16.3% (n = 152) selected headline because it
was “more interesting.” The primary reason that the smallest proportion of the sample selected a
headline (14.7%, n = 137) was because the headline was “more personalized” for them. The
option to check “other” was used less than ten times.
Table 2.
Number and percentage of reasons for headline selection
Reason for Selection
n
%
More Informative
449
48.0
Raised Curiosity
197
21.1
More Interesting
152
16.3
More Personalized
137
14.7
Further comparison of the mean percentages and t-test results of the reasons for selection
by level of informativeness (Table 3), the mean of 61.6 % (SD = 4.7%) selected a high-
information headline because it was more informative compared to 22.7 % (SD = 7.1%) selecting
a low information headline because it was more informative. An independent samples t test
indicated this difference was statistically significant (t=12.1, p = <.001). At the same time for
headlines that raised curiosity, the majority (M = 34.8%, SD = 7.9%), selected a low-information
headline while a lower percentage of participants (M = 13.8%, SD = 2.4%) chose a highly

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
11
informative headline because it raised curiosity yielding a significant t test (t = -6.7, p = <001).
Low-information headlines appeared to be more interesting (M = 23.2%, SD = 7.5%) as
compared to high-information headlines (M = 12.6 %, SD = 3.4%) and this difference was
significant (t = -3.4, p = < .001).
Finally, although the reason of a headline being more personalized was indicated as the
least of the four reasons for selection, more users selected low informative headlines (M = 19.4,
SD = 5.3) for personalization than those selecting highly informative headlines (M = 12.0, SD =
4.2) and this difference was also significant, (t = - 3.4, p = .01).
Table 3.
Mean percentages and t-test results for selecting high versus low information headlines
Reason
High
Informativeness
Low
Informativeness
M
SD
M
SD
t
Sig
More Informative
61.6
4.7
22.7
7.1
12.1
<.001 **
Raised Curiosity
13.8
2.4
34.8
7.9
-6.7
<.001 **
More Interesting
12.6
3.4
23.2
7.5
-3.4
<.001 **
More Personalized
12.0
4.2
19.4
5.3
-2.9
.01 *
*p = <. 05 **p = < .001
Headline interactivity
Table 4 details interactivity with headlines by levels of informativeness with slightly
more “likes” for low information headlines (M = 39.4 %, SD = 8.8%) compared to “likes” for
high information headlines (M = 33.0 %, SD = 6.0%) but this difference was not statistically
significant. In second place for interactivity, more (M = 38.6 %, SD = 7.4%) sharing was

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
12
anticipated for low information headlines compared to high information headlines (M = 28.5%,
SD = 7.3%) and this difference was significant (t = -2.6, p = .02).
Finally, more participants said they would more likely comment on low informative
headlines (M = 13.5 %, SD = 6.5%) than high informative headlines (M = 8.5%, SD = 3.7%) but
the difference was not significant.
Table 4.
Means and t-tests comparing reasons for interactions with selected headlines.
Activity
High
Informativeness
Low
Informativeness
M
SD
M
t
Sig.
Like
33.0
6.0
39.4
8.8
-1.6
0.13
Share
28.5
7.3
38.6
7.4
-2.6
0.02 *
Comment
8.5
3.7
13.5
6.5
-1.8
0.09
DISCUSSION
The goals of this online exploratory story were to measure any significant differences in
the selection of highly informative versus low information headlines, the reasons for selecting a
headline, and user interactivity with selected headlines.
The results reported here raise some interesting points about how quickly user interest is
determined for health news and why. To start, many news producers have believed – and some
still believe – that the key to enticing users to click a headline is to “tease” the user with only
limited information (so called “clickbait). Results reported across eight different health news

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
13
stories provide strong evidence that low information “teasing” headlines do not guarantee
selection. In fact, the opposite outcomes were statistically significant.
Focusing on varying levels of information, headlines with the most information were
selected significantly more often than the headlines with less information.
As for the reasons for selection, overall, nearly half of the participants selected a headline
because it was more informative with about 16% selecting because it was “more interesting.”
On average, although participants were able to select one of the five reasons why they selected a
headline, most (62%) selecting highly informative headlines specifically said it was because
more details were provided. Significantly fewer participants (about 14%) selected a highly
informative headline because it “raised curiosity.”
Conversely, 35% of the participants who selected the low information headline in the pair
said it was because the headline raised curiosity. Only 23% of those selecting the low
information headline said it was more informative.
Although participants could select one or more interactive responses to like, share and/or
comment on content related to the selected headline, only sharing produced a significant
difference between headlines with more intentions to share low information headlines. One
possible reason is that the novel curiosity generated by some low information headlines triggered
a desire to share such novelty with others, as opposed to high information headlines that are just
typically declarative sentences that provide more details. While slightly more “likes” and
commenting were anticipated for low information headlines, these were not statistically different
from the likes and comments for high information headlines. Combined, these results might
suggest that headlines containing less information are more likely to generate user interactivity,
but more evidence is needed before such a claim could be generalized.

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
14
At the same time, participants in this study reported that low information headlines (23%)
were “more interesting” than high information headlines (13%). While one could only speculate
why this opposite outcome occurred, one possible explanation is that the reason that a headline
“raised curiosity” might have been confusingly similar to “was more interesting.” Future
research should define such terms.
The findings of this study have theoretical implications. The theory of incidental
exposure suggests that getting exposed to news is a by-product of users’ online activities. Users
are spending a large amount of time navigating social media to meet their need for information.
News organizations also offer content on social media to garner the attention of the users. That
means news organizations are showcasing their content with a view to attracting the attention of
these floating audiences, who do not actively seek news with prior intention. But when they find
any interesting news, they tend to read it, share and comment on it.
Similarly, the results of this study showed that users are interested to like, share, and
comment on health news when they find it on social media. Participants of this study tend to be
interactive both in the contexts of high and low informative headlines. This suggests that news
managers could write the headlines considering the levels of informativeness and present it to
readers via digital spaces. This ultimately would help news organizations to reach out to wider
audiences. In this way, news organizations become able to present reliable health news to the
audiences, which also help combat the spread of health-related misinformation. So, it can be
assumed that offering health news of legacy news media to the wider audiences would likely act
as an informational tool.
Limitations

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
15
Despite several statistically significant outcomes, it is possible that more detailed
differences would occur with a larger sample size. Given that data collection was unexpectedly
interrupted by the COVID-19 crisis, forcing instruction from classrooms to online, the intended
larger sample was not collected. On the other hand, the level of significance for many of the
measures suggests that the effects of varying headline informativeness may be large enough to
support smaller samples.
A second limitation is that this study focused only on health news headlines. The
researchers cannot make any claim to generalize these results to other types of headlines or
other content. Future research should apply similar coding and measures to see if other types
of headlines would produce similar results.
Finally, even within the health news genre, the headlines from only eight health news
stories served as stimuli. Future studies that explore health news should continue to expand
the variety of health topics to see if also varying the subject of health issues and/or ailments
would produce the same effects of varying the levels of headline informativeness.
Conclusion
Overall, this exploratory experimental research testing relatively small differences in the
amounts of information presented in 8 pairs of headlines produced an impressive collection of
systematic and significant differences worth noting.
At least in this case, there was an overwhelming preference for health news headlines
providing more information because they, in fact, provided the user with more details. These
results might suggest that digital users who are increasingly busy, distracted and/or checking
news on their smaller mobile screens may prefer easy-to-read and understand headlines with
more information that instantly convey useful information in a relatively short period of time.

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
16
This would be especially true for scanning users as opposed to those who have more available
time to read a health news story that is much longer than a headline.
If true, such justification challenges the assumption that less informative “teasing”
headlines are more effective in generating more clicks to news stories. Clearly, some users will
always click less informative “teasing” headlines regardless, but these results suggest that
perhaps even more users would click on more headlines when they provide more information as
opposed to less information.

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
17
REFERENCES
Baum, M. A. (2011). Soft news goes to war: Public opinion and American foreign policy in the
new media age. Princeton University Press.
Boczkowski P, Mitchelstein E, and Matassi M (2017). Incidental news: How young people
consume news on social media. Annual meeting of the Hawaii International Conference
on System Sciences, Hawaii, January 3-7.
Boland, G. (2017). How to Write Attention-Grabbing Headlines in 2017. Newswhip. Retrieved
on March 18, 2020 from
http://web.archive.org/web/20170221125611/https://www.newswhip.com/2017/01/top-
publisher-headlines-2015-vs-2017/
Dor, D. (2003). On newspaper headlines as relevance optimizers. Journal of Pragmatics, 35(5),
695-721.
Downs, A. (1957). An Economic Theory of Democracy. New York: Harper.
Fletcher, R., & Nielsen, R. K. (2018). Are people incidentally exposed to news on social media?
A comparative analysis. New media & society, 20(7), 2450-2468.
Forster, K. (2017). Revealed: How dangerous fake health news conquered Facebook. The
Independent. Retrieved on March 16, 2020 from https://www.independent.co.uk/life-
style/health-and-families/health-news/fake-news-health-facebook-cruel-damaging-social-
media-mike-adams-natural-health-ranger-conspiracy-a7498201.html
Funk, C., Gottfried, J., & Mitchell, A., (2017). Most Americans express curiosity in science
news, but a minority are active science news consumers. Pew Research Center. Retrieved
on March 18, 2020 from https://www.journalism.org/2017/09/20/most-americans-
express-curiosity-in-science-news-but-a-minority-are-active-science-news-consumers/

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
18
Geer, J. G., & Kahn, K. F. (1993). Grabbing attention: An experimental investigation of
headlines during campaigns. Political Communication, 10(2), 175-191.
Growney, C. M., & Hess, T. M. (2019, March 4). The Influence of Mood Versus Relevant Self-
Perceptions in Older Adults’ Interest in Negative Health-Related Information.
Psychology and Aging. Advance online publication.
http://dx.doi.org/10.1037/pag0000333
Herzog, H. (1941). On borrowed experience: An analysis of listening to daytime
sketches. Zeitschrift für Sozialforschung, 9(1), 65-95.
Ifantidou, E. (2009). Newspaper headlines and relevance: Ad hoc concepts in ad hoc
contexts. Journal of Pragmatics, 41(4), 699-720.
Jenkins H (2006) Convergence Culture: Where Old and New Media Collide. New York: New
York University Press.
Ksiazek, T. B., Peer, L., & Lessard, K. (2016). User engagement with online news:
Conceptualizing interactivity and exploring the relationship between online news videos
and user comments. New media & society, 18(3), 502-520.
Lee J (2009) Incidental exposure to news: limiting fragmentation in the new media environment.
Unpublished Doctoral Thesis, University of Texas at Austin, Austin, TX.
Lopez, C., Prince, V., & Roche, M. (2014). How can catchy titles be generated without loss of
informativeness?. Expert systems with applications, 41(4), 1051-1062.
Mitchell A, Gottfriend J, Shearer E, et al. (2017) How Americans encounter, recall and act upon
digital news. Pew Research Center. Available at:
http://www.journalism.org/2017/02/09/how-americans-encounter-recall-and-act-upon-
digital-news/ (accessed March 2020).

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
19
Napoli P (2011) Audience Evolution: New Technologies and the Transformation of Media
Audiences. New York: Columbia University Press.
Nguyen, A. (2008). The contribution of online news attributes to its diffusion: An empirical
exploration based on a proposed theoretical model for the micro-process of online news
adoption/use. First Monday, 13(4).
Raphael, R. (2019). A shockingly large majority of health news shared on Facebook is fake or
misleading. Fast Company. Retrieved on March 15, 2020 from
https://www.fastcompany.com/90301427/a-shockingly-large-majority-of-health-news-
shared-on-facebook-is-fake
Shapiro, N. (2018). The Fake News Epidemic in Health. Daily Beast. Retrieved on March 14,
2020 from https://www.thedailybeast.com/the-fake-news-epidemic-in-health?ref=author
Tewksbury, D., Weaver, A. J., & Maddex, B. D. (2001). Accidentally informed: Incidental news
exposure on the World Wide Web. Journalism & Mass Communication Quarterly, 78(3),
533-554.
Thurman N (2008) Forums for citizen journalists? Adoption of user generated content initiatives
by online news media. New Media & Society 10(1): 139–157.
Valeriani A and Vaccari C (2016) Accidental exposure to politics on social media as online
participation equalizer in Germany, Italy, and the United Kingdom. New Media & Society
18: 1857–1874.
Williams, R. (2003). Television: Technology and cultural form. Psychology Press.
Yadamsuren, B., & Erdelez, S. (2010). Incidental exposure to online news. Proceedings of the
American Society for Information Science and Technology, 47(1), 1-8.

RUNNING HEAD: INFORMATION IN HEALTH HEADLINES
20
Zukin, C., & Snyder, R. (1984). Passive learning: When the media environment is the message.
Public Opinion Quarterly, 48(3), 629-638.