Twitter Makes It Worse: Political Journalists, Gendered Echo Chambers, and The Amplification of Gender Bias
Twitter Makes It Worse: Political Journalists, Gendered Echo Chambers, and The Amplification of Gender Bias
research-article2018
HIJXXX10.1177/1940161218781254The International Journal of Press/PoliticsUsher et al.
Research Article
The International Journal of Press/Politics
1–21
Twitter Makes It © The Author(s) 2018
Reprints and permissions:
Worse: Political Journalists, sagepub.com/journalsPermissions.nav
DOI: 10.1177/1940161218781254
https://doi.org/10.1177/1940161218781254
Gendered Echo Chambers, journals.sagepub.com/home/hij
Abstract
Given both the historical legacy and the contemporary awareness about gender
inequity in journalism and politics as well as the increasing importance of Twitter in
political communication, this article considers whether the platform makes some of
the existing gender bias against women in political journalism even worse. Using a
framework that characterizes journalists’ Twitter behavior in terms of the dimensions
of their peer-to-peer relationships and a comprehensive sample of permanently
credentialed journalists for the U.S. Congress, substantial evidence of gender bias
beyond existing inequities emerges. Most alarming is that male journalists amplify
and engage male peers almost exclusively, while female journalists tend to engage
most with each other. The significant support for claims of gender asymmetry as
well as evidence of gender silos are findings that not only underscore the importance
of further research but also suggest overarching consequences for the structure of
contemporary political communication.
Keywords
political journalism, gender, Twitter, Washington journalism, beltway journalism,
women in journalism
Corresponding Author:
Nikki Usher, University of Illinois, College of Media, 119 Gregory Hall, 810 S. Wright Street, Urbana, IL,
61801 USA.
Email: [email protected]
2 The International Journal of Press/Politics 00(0)
As with other creative and knowledge industries, problematic gender dynamics have
come to the forefront in journalism amid extensive news coverage about endemic sex-
ual harassment. Journalist-on-journalist sexual harassment was reported in the United
States; in Washington, D.C., top political journalists at MSNBC, Vox, The New Republic,
National Public Radio (NPR), Fox News, and elsewhere resigned amid allegations. A
“shitty media men” list circulated, listing powerful male journalists at top news organi-
zations. Other journalist-on-journalist sexual harassment stories surfaced in Sweden,
France, Japan, Italy, Pakistan, the United Kingdom, and Singapore, often with female
journalists reporting abuse but unable to hold male journalists accountable (Allsop and
Ho, 2017).
Gender asymmetry in journalism has been observed internationally across employ-
ment patterns, sourcing behaviors, and news decision making: in the United Kingdom
and Ireland (Ross et al. 2018); in Australia (Hannis and Strong 2007); in Hong Kong
(Tsui and Lee 2012); in Germany, Senegal, and Norway (Van Zoonen 2002); and
beyond. According to the 2016 American Society of News Editors Diversity Survey,
across 737 U.S. news organizations (mostly newspapers, along with some digital-
native publishers), women make up just a third of journalists. A study of the coverage
patterns of twenty leading U.S. news outlets showed that female journalists report and
produce less news than men; 37.7 percent to 62.3 percent across all platforms (Women’s
Media Center 2017).
Thus far, research on gender dynamics on social media has focused on political
candidates (cf. McGregor and Mourão 2016), with scant attention by political com-
munication scholars to these questions in political journalism. Given the pervasive
evidence of off-line gender disparities in journalism, this article interrogates how gen-
der dynamics impact the way Washington, D.C., political journalists (“beltway jour-
nalists”) relate to each other on Twitter, an all-important platform critical to how
beltway journalists do their work. An overarching question orients this research: Do
the off-line gender disparities in beltway journalism play out on Twitter, too? If so,
how? On Twitter, political journalists tweet to and about other political journalists
more than they engage with any other type of user on Twitter (Molyneux 2015;
Molyneux and Mourão 2017; Mourão et al. 2016).
Thus, rather than examining journalists’ overall Twitter activity, we examine
whether gender imbalances are present in beltway journalists’ peer-to-peer relation-
ships on Twitter—or how journalists use the platform to legitimate, amplify, and
engage each other. We focus on the United States here and develop a new approach to
analyzing how political journalists use Twitter that advances our understanding of the
substantially important gender imbalances we document in U.S. political journalism—
but which are likely also present elsewhere. As such, we provide a template that we
hope will empower further research internationally to advance our understanding of
cross-national differences in political communication, news production, and the use of
social media by journalists.
To explore these questions, we created a sample consisting of all beltway journal-
ists with active Twitter accounts who are credentialed to cover the U.S. Congress. To
be included in our sample, these journalists must have been living and working in the
Usher et al. 3
Washington, D.C., area (referred to colloquially as “the beltway” because of the circu-
lar freeway that wraps around Washington and its immediate suburbs). While research
on Twitter has often been critiqued as failing to be an adequate stand-in for a larger
non-Twitter using population, in the case of political journalists, Twitter is absolutely
integral to their daily work practices and reputational success in the United States
(Hamby 2013; Kreiss 2016), but also increasingly in Europe (Nuernbergk 2016) and
Westminster democracies (Hanusch 2017).
Our normative position must be established at the outset: While journalism research
itself is mixed on the role of the gender of journalists in shaping news coverage (Hayes
and Lawless 2015), we believe that it is a valuable social outcome to have more women
in journalism, especially in positions of authority and influence. This is not just a ques-
tion of equity: Greater representation of women in newsrooms may well change how
female journalists are treated by their colleagues, their advancement, story selection,
framing, and sourcing patterns (Mills 1997). In our sample, the number of male belt-
way journalists outnumbered female beltway journalists both off-line and on Twitter.
Even if Twitter just reflected these existing gender asymmetries, this would already be
a problem. But if Twitter data suggested even worse gender asymmetries, given
Twitter’s critical role in the news production process, it might be actually augmenting
gender inequality across political communication in the United States. Thus, the arti-
cle first discusses why the Twitter dynamics of political journalists are essential to
understanding political communication, documents issues facing female journalists
and female political journalists, and reviews what we know about gender, journalism,
and Twitter. It then moves to explicate our research questions, methods, and findings.
Overall, we find support that Twitter reflects existing gender disparities in the beltway
(and, indeed, may amplify them)—male journalists “win” and win big on Twitter
across most measures; moreover, men appear almost to self-segregate, while women
only do so in limited circumstances.
accounts across twenty-one U.S. news organizations, finding journalists were less
likely to follow women as a whole and women were less likely to be retweeted, but did
not discuss journalist-to-journalist behavior.
Other smaller scale studies indicate divergent gender patterns among journalists; a
study of local journalists’ tweets during a power outage found men who were more
likely to send tweets served a one-way broadcast role, rather than engaging with audi-
ences (Nee and Rebecca 2015). Research on female Indian journalists highlighted the
toxic nature of the platform (Gudipaty 2017), while a study of U.S. foreign correspon-
dents found limited use of Twitter and no gender difference (Cozma and Chen 2013).
Using a convenience sample of forty-five political journalists, Parmelee et al. (2017)
found male political journalists were more likely to engage in replies, although the
sample makes it hard to generalize across aggregate behavior. Other research revealed
female journalists were more likely than men to link outside their own organizations
and tweeted more about their personal lives (Lasorsa 2012)—Twitter practices that
may subject female journalists already vulnerable to online sexual harassment to fur-
ther abuse (Adams 2017; Mantilla 2013).
Research Questions
We, thus, approach our analysis of gender dynamics on beltway journalism Twitter
with three research questions:
Research Question 1 (RQ1): Is there gender asymmetry in the way that beltway
journalists legitimate each other on Twitter?
Research Question 2 (RQ2): Is there gender asymmetry in the way that beltway
journalists engage with other beltway journalists on Twitter?
Research Question 3 (RQ3): Is there gender asymmetry in the way that beltway
journalists amplify other beltway journalists on Twitter?
Research questions guide our study rather than hypotheses, in part, because we take
an inductive approach and, in part, because outliers in our data make verifiable hypoth-
eses across the entirety of the data set less meaningful. This approach is used in other
research on political journalism and Twitter (cf. Lawrence et al. 2014). Our first ques-
tion considers whether Twitter cues about a journalists’ cultural capital might be gen-
dered, thus, possibly impacting how journalists assess the legitimacy of other
journalists. The second and third research questions reflect how journalists treat each
other—probing whether the peer-to-peer dynamics on Twitter among beltway journal-
ists differ based on gender. There is little guidance or agreement in the political com-
munication literature about which specific Twitter behaviors should be used as primary
indicators to assess community discourse. In fact, Freelon et al. (2018) argue it is
important “not to simply assume without justification that particular traces signify
particular theoretical constructs” (p. 993)—whereby certain trace-based metrics may
be considered more or less important based on the underlying theoretical questions
and community context.
Usher et al. 7
Sample
We created a purposive sample of 2,292 nonprotected Twitter accounts of beltway
journalists credentialed to cover Congress (1,299 men or 56.7 percent of the sample;
993 women or 43.3 percent of the sample). While we collected tweets from Twitter’s
Timeline Application Programming Interface (API) for a broader time period, we lim-
ited our data set to tweets posted from June 1 to August 1, 2017. The total data set of
817,136 reflects every nondeleted tweet of these journalists, but does not include
tweets from other accounts that reference the journalists in our sample.1 We drew from
the list of credentialed congressional correspondents found in the Congressional
Directory for the 114th Congress of the United States. The Standing Committee of
Correspondents, journalists elected by their peers, controls the overall congressional
credentialing process. To be credentialed, a journalist must permanently reside in the
D.C. area, work full-time as a journalist, and be “editorially independent of any insti-
tution, foundation or interest group that lobbies the federal government, or that is not
principally a general news organization” (Congressional Directory 2016: 976). The
White House has its own accreditation process and does not make its list public; how-
ever, a prerequisite of membership for The White House Correspondents’ Association,
a journalist-run association, is credentialing by the congressional committee. Our
sample does not include every political reporter covering politics in Washington (for
example, Maggie Haberman of the New York Times, who lives in New York). As the
official Congressional Directory is only updated every two years, newly credentialed
journalists are not in our sample. Nonetheless, the directory contains a substantial
representation of political journalists who live and work in the beltway.
To arrive at the final list of 2,292 beltway journalists, we eliminated some accounts
from a broader list of 2,546 American/Anglo media journalists. To do so, we identified
journalists with active Twitter accounts and updated their entries, eliminating journal-
ists who were identified as no longer working for a news organization or in Washington,
D.C. This list-building process took two months and was completed on May 31, 2017.
Over these two months, some journalists might have changed jobs or left journalism;
however, we believe including the tweeting behavior of these journalists would not
substantially affect our results.
Our sample is unique: It is a purposive account-driven data set of journalists who
share a definably similar characteristic: They are elite political journalists doing work
considered worthy of permanent credentials to cover the United States’ national legis-
lative body. By collecting the tweets of a specific set of users, this study is different
from others that rely on keywords or hashtags. Our sample includes a variety of types
of news media not distinguished in any meaningful way by the Congressional
Directory, as in a cross-media era, the different types of political media are intermin-
gled (niche, local, national, radio, TV, digital native, etc.). Moreover, young journalists
can start right out of college working at top-flight national news outlets, niche outlets
play an important role in the news ecosystem, and networking across beats and news
outlets is a well-established Washington phenomena. For these reasons, our sample is
more similar than different because of the peculiarities of the beltway and is as “apples
8 The International Journal of Press/Politics 00(0)
to apples” as one can get for political journalists; while some might have greater cul-
tural capital or more experience, as a whole, these are journalists in the elite strato-
sphere of political reporting.
Data Collection
From June 1, 2017 to August 1, 2017, we collected tweets, followers, and profile informa-
tion from a final list of 2,292 nonprotected, active Twitter accounts of journalists (journal-
ists who did not tweet during the collection period were removed). Collection was
performed using Social Feed Manager (George Washington University Libraries 2016),
an open-source software that harvests data from Twitter, Tumblr, Flickr, and Sina Wiebo,
and for this project, relied on Twitter’s Timeline API. During the data collection period, a
small number of accounts were deleted or became protected and were removed from the
sample. A small number of accounts changed handles (the @ identifier for an account),
but because Social Feed Manager relies on user IDs, this did not affect collection.
Measures
We assess legitimation, amplification, and engagement patterns of beltway journalists
on Twitter through measures derived from our research questions, extant theory, and
context.
Gender
Our most critical measure is gender itself. Each Twitter account was manually coded
for gender and assessed via normative social constructions of gender: by name, by
gender presentation in profile photos, and by secondary information such as Twitter
bios and self-identification in the journalist’s twenty most recent tweets. Gender here
is constructed as a binary, and it is acknowledged that ascribing gender raises issues.
Power Users
The top twenty-five users in the ranking for each Twitter activity are defined here as
“power users.” To determine them, the users were ranked for each of the Twitter activi-
ties (mentions, retweets/quotes, replies, and following) by total count of the activity.
This gives us more detailed insight into who the biggest players are in beltway journal-
ism Twitter. Power users are sometimes called influencers because they may dispro-
portionately impact the amplification patterns of a tweet or its underlying information
(Bakshy et al. 2011), but in our case, “power user” is more appropriate because it
examines a wider range of Twitter activities and their possible impact.
Verification status is the blue check next to an account. Twitter decides whether some-
one’s account can be verified, reserved the check mark for who the platform dubs a
“public figure.” News organizations, however, can help staff get verified by sending
lists of accounts to Twitter.
Total follower numbers are often considered a low-value indicator, what Cha et al.
(2010) describe as the “the million follower fallacy.” There is increasing evidence that
follower numbers have significant noise, from bots to users who may inflate numbers
by purchasing followers. But the widespread perception among Twitter users is that
follower numbers connote influence (Cha et al. 2010), and so the cultural capital
inferred upon journalists by this external validation is nonetheless useful.
Mentions by other journalists also signal legitimacy because of their particular con-
versational function: a “shout out” to another journalist, which may be particularly
important if the journalist doing the “shout out” has many followers herself. The more
mentions a peer has, particularly by users perceived as having higher cultural capital
on the platform, the more important that journalist may be taken to be in the insular
world of beltway journalist Twitter. Of course, one can be mentioned on Twitter nega-
tively as well as positively, but tone is beyond the scope of this study.
Original tweets are ways to brand oneself, to say something new, and to initiate a
conversation. When it comes to measures of legitimacy on the platform, original
tweets are the only factor that the journalist herself gets to control.
Note. Table shows activity of power users for activity with significant results.
a retweet with a comment, in our analysis, we consider them similar enough actions to
treat as one category; below, the word “retweet” should be taken to include “retweet/
quotes.” (As users often note, Twitter retweets do not always necessarily constitute
endorsements.)
Results
Several layers of tests across multiple measures of gender parity suggest a consistent,
though not unanimous, imbalance between male and female Washington-based politi-
cal journalists on Twitter. The central starting point for our analysis is the breakdown
of the sample as a whole—of the 2,292 nonprotected Twitter accounts of beltway
journalists credentialed to cover Congress who tweeted during the collection period,
there are 1,299 men and 993 women, for a statistically significant gender split of 56.7
percent male journalists to 43.3 percent female journalists (p < .001). While the split
as a whole favors men, when a measure matches the split, it suggests that the bias
reflects existing inequity; a measure that does not match the split suggests further
gender bias (assuming a significant result). We used a chi-square goodness-of-fit test
to calculate whether the gender differences on any specific activity/measure were sig-
nificant when compared with the split, while the average account-level Twitter mea-
sures were compared by gender using a standard difference-of-means test. We approach
our analysis in three key ways to examine whether beltway journalists differ by gen-
der: how they are legitimated, engaged, and amplified. To look at these dynamics in a
more descriptive way, we looked at the gender dynamics of the top twenty-five power
users (see Table 1).
Usher et al. 11
Note. Tests are chi-square goodness of fit comparisons (Degrees of Freedom = 1) unless indicated by −
x,
when mean difference test was conducted. Verification data represent the share of each gender that is
verified.
*p < .05. ***p < .001.
Legitimation
By legitimation, we mean the account characteristics of beltway journalists that signal
cultural capital on Twitter (Barnard 2014)—indicators of a journalists’ legitimacy as a
member of the in-group. We considered four key characteristics by gender: verifica-
tion status, journalists’ follower count, journalists’ mention counts by other journalists,
and journalists’ original tweets.
First, we considered verification status, which we found to be statistically signifi-
cant: Male political journalists are slightly more likely to be verified than female jour-
nalists, with the little blue check mark of authenticity conferring legitimacy in a way
suggestive of gender bias.
Second, we considered the possible impact of a journalists’ number of total follow-
ers. We find male journalists are followed at roughly twice the rate of women; male
journalists were, on average, followed by 20,181.31 total accounts, while female journal-
ists were, on average, followed by 11,609.53 total accounts (8,517.786; p < .001, x−).
If journalists are making an assessment about which of their peers is worth paying
attention to on Twitter, this huge gender asymmetry in aggregate followers may well
have a disproportionate impact in the perceived legitimacy of their peers.
Third, we looked at how often a journalist was mentioned by another journalist. We
argue that mention counts provide an important marker of cultural capital on the plat-
form. On average, these mention patterns by gender of journalists do not differ in a
significant way (see Table 2), a hopeful sign: Male and female journalists are giving
“shout outs” in a way that does not suggest gender bias.
Fourth, we looked at original tweets. For journalists assessing the cultural capital of
a peer journalist on Twitter, a key factor may be whether a journalist is tweeting inter-
esting content. These original tweets (as opposed to retweets or replies) constitute 28.6
percent of all of the tweets in our sample and, thus, play an important role in overall
dynamics. We note that, on average, male journalists craft more original tweets than
female journalists: Men tweeted 115.993 original tweets, while women tweeted 83.838
tweets (p < .001) during the eight-week period studied; in aggregate, male beltway
journalists do more original tweeting than female beltway journalists (see Table 2).
12 The International Journal of Press/Politics 00(0)
Engagement
To look at engagement, we considered two key indicators. First, we looked at follow-
ing patterns by journalists of other journalists. This helps us understand how journal-
ists are paying attention to each other, and explore whether this differs by gender.
Then, we looked at reply patterns, which provides insight into how journalists respond
to each other.
Following. We find statistical support that there is gender asymmetry among whom
beltway journalists choose to follow (see Table 3). On average, beltway journalists
(taken as the entire sample of male and female journalists) are more likely to follow
male journalists than female journalists; male journalists follow male journalists more,
but female journalists also follow male journalists more.
The top twenty-five beltway journalists followed by other beltway journalists pro-
vides additional descriptive nuance to this asymmetry: The list only includes four
women (see Table 1). The first woman in the top twenty-five most-followed list does
not even make it into the top ten; she is Andrea Mitchell, who has covered politics for
Usher et al. 13
Note. Tests are chi-square goodness of fit comparisons (Degrees of Freedom = 1), unless indicated by −
x,
when mean difference test was conducted.
*p < .05. **p < .005. ***p < .001.
fifty years for every major U.S. broadcast network, CNN, and MSNBC. These follow-
ing patterns suggest a systemic problem: If female journalists are not followed to the
same degree as male journalists by other beltway journalists, their capacity to be seen
and engaged with is much lower.
Reply Patterns. We examined whether beltway journalists were more likely to reply to
male journalists than to female journalists. For context, when beltway journalists reply
to other users, almost a third of these replies are to other beltway journalists’ accounts.
As a whole, when beltway journalists reply to other beltway journalists, 76.5 percent
of those replies are to male journalists (p < .001; see Table 3). The list of the top
twenty-five most-replied-to beltway journalists by other beltway journalists is also
suggestive: It includes only four women (see Table 1); none are in the top ten.
Another indicator of gender asymmetry is whether a journalist’s gender affects who
they reply to; for example, does being a male journalist make it more likely that the
journalist will reply to other men? We find that when male journalists reply to other
beltway journalists, they reply to another male journalist 91.5 percent of the time (p <
.001). That means male journalists only reply to female journalists 8.5 percent of the
time. In contrast, we find that female journalists are most likely to respond to other
female journalists. When female journalists reply to other beltway journalists, they
reply to female journalists 72.1 percent of the time (p < .001). We see the starkest
evidence of gender asymmetry at play when we look at reply patterns by journalists—
in fact, there are arguably gender silos with men existing in their own bubble replying
to other men, women existing in their own bubble replying to other women.
Amplification
We were interested in understanding whether amplification patterns among beltway
journalists were gendered. The total retweets and quote category represents the most
14 The International Journal of Press/Politics 00(0)
substantial portion of our data—56 percent of all tweets in the data set (42.3 percent
retweets, 13.7 percent quotes); 25.6 percent of these retweets are journalists retweet-
ing other journalists.
First, we considered whether beltway journalists as a whole retweeted male journal-
ists more often than they retweeted female journalists. We found that when beltway
journalists retweeted other journalists, they retweeted men 68.9 percent of the time and
women 31.1 percent of the time (p < .001; see Table 4). Of the top twenty-five most
retweeted journalists by fellow journalists, only five were women. In this case, a woman
did break into the top ten—Seung Min Kim, at the time Politico’s highest ranking con-
gressional reporter, was the sixth most retweeted beltway journalist by her peers.
We then assessed whether the gender of the journalist made a difference as to
whether he or she was more likely to retweet a man or a woman. Based on our reply
patterns, which showed strong evidence of gender self-segregating, we thought we
might see the same pattern in retweets. We found that male journalists, indeed, were
more likely to retweet other male journalists—in fact, when male journalists retweet,
they retweet men almost three times more often than they retweet women (see Table
4). We then examined if female journalists were more likely to retweet female jour-
nalists or male journalists. Like male journalists, female journalists also retweeted
male journalists significantly more often than female journalists (56.9 percent to
40.4 percent, p < .001). Thus, men are siloed in their retweet patterns, whereas women
amplify male voices more than women’s. Overall, these retweet patterns provide evi-
dence that female beltway journalists face an uphill battle in being amplified by their
peers.
Discussion
Our research found significant indicators that gender bias among beltway journalists
is present on Twitter. We knew from the outset that there was, indeed, a statistically
significant gender imbalance between the number of male and female beltway journal-
ists on Twitter, already a problem given our normative concern. What would be more
concerning was whether these asymmetries were, in fact, somehow worse on Twitter,
which they, indeed, were: Results are suggestive not just of male beltway journalists
Usher et al. 15
dominating on Twitter, but also of men engaging with male peers and leaving women
out of the picture. In trying to make sense of our findings, we posit that the nature of
the platform itself, the hostile and toxic environment women face online (Mantilla
2013), and implicit gender bias sets women up to be less represented on the platform,
and may well create an even greater structural disadvantage for female journalists,
given how this platform is so critical to success in beltway journalism.
Male journalists’ accounts have more cues that legitimate them in the eyes of other
beltway journalists. Men are more likely to have the all-important sign-off as a veri-
fied Twitter account, a sign they are a “public figure” in the eyes of the tech company
(or the news organization that may have submitted their account for verification), they
have more followers as a whole and they tweet more, cues for their cultural capital on
the platform. When it comes to engagement patterns, men reply to men an astonishing
percentage of the time, and both men and women follow more male beltway journal-
ists. Amplification patterns advantage men more than women, with the vast majority
of retweets featuring male journalists.
Given the “small world” effect of Web dynamics in political communication
(Jürgens et al. 2011), some journalists have outsized capacity to promote other journal-
ists, and our data suggest that male journalists will get heard more than women. Our
work looked at the aggregate behavior of our sample as well as among power users,
giving an added dimension of not just what journalists on Twitter do, but also insight
into gender disparities among the most influential political journalists on Twitter. With
the exception of female journalists’ reply activity, for every other category of Twitter
activity (with statistical significance), eighteen or more men were featured in the top
twenty-five power users list.
A journalist has different levels of agency when it comes to the factors that affect
whether peers might legitimate her: verification status is largely out of her hands, as is
the external validation accorded by her aggregate number of Twitter followers. The
frequency to which other journalists give her a “shout out” may be determined by
innumerable and intangible reasons beyond her control. However, original tweets do
seem to matter, and we hypothesize may have an even larger impact than our results
show: original tweets are the wellspring of engagement and amplification by other
Twitter users, and if women tweet less, they will be heard less.
We were most surprised to see the extent to which some of this disparity suggests a
gendered echo chamber, especially when it comes to male journalists’ engaging and
amplifying each other. Although replies by beltway journalists to other beltway jour-
nalists represent a small portion of overall tweets, the patterns were eye-popping: men
replying to men 91.5% of the time; women replying to other women 72.1 percent of
the time, dynamics also demonstrated among our power users (see Table 1). Retweet
patterns suggest that men live in a gendered echo chamber that promotes other male
journalists at the expense of female ones: Men retweet other men almost three times
more often than they retweet women. In contrast, women still retweet male journalists
more than they retweet female journalists. As Freelon, McIlwain, and Clark (2016)
observe in their #blacklivesmatter research:
16 The International Journal of Press/Politics 00(0)
replies and retweets may not always be endorsements but they do partially determine
which participants are seen as experts and leaders. Put more directly, they confer power,
and the women in our dataset have not received a proportionate share. (p. 14)
The same could be said for our data set. At times, female journalists confer power
among other female journalists, but this does not change the overall pattern of male
beltway journalists’ domination on Twitter. The underproduction of original tweets of
women compared with men may make it harder for both women and men to combat
these gendered dynamics; one can only reply or retweet if one has something to
respond to. One possible reason for this disparity in original tweets may be that women
do more emotional labor at home and at work (Stevens et al. 2001), so may have less
time to pay attention to Twitter, or perhaps women see a diminished utility from time
spent on an emotionally burdensome platform relative to their other work responsibili-
ties (Megarry 2014).
Of particular interest is the strong preference women have for replying to each
other. Perhaps women are engaging with each other in a more supportive way, and so
some of the anxiety, not to mention the threats and harassment that women face on
Twitter, may be mitigated when female journalists talk to each other. This may also be
an intentional and conscious effort on the part of female journalists to attempt to create
a sliver of a supportive community on a digital platform that is known for its trolling
(Adams 2017). Yet, women follow more men and retweet more men, which might be
explained by a variety of reasons. Women may follow the same cues that men do when
assessing an account’s perceived value on the platform. As our data do not distinguish
sentiment, it may be that men are more willing to “go negative” or challenge other men
on the platform, while women turn away from this tenor of activity, or in the case of
retweet/quote tweets, may be more comfortable critiquing men than women.
Our research did not account for the intersection of gender and the prestige of a
media outlet, nor did it account for dimensions of partisanship or age. However, these
journalists are all credentialed to work in the same physical setting, the press galleries
of the House and Senate, and they intermingle in the same social settings (The Press
Club, award dinners, bars, etc.), and young people are often vaulted into some of the
most prestigious jobs in beltway journalism. We contend these journalists are arguably
the most uniform sample of U.S. political journalists analyzed thus far.
Further research is warranted. Comparative work across credentialed journalists
may reveal the role national culture plays in these dynamics we observed. Secondary
analysis would be instructive to see how much men link to other men in these original
tweets; if most original tweets do have some specific reference to one’s original jour-
nalism content, reducing byline disparities may help overcome some gender issues.
Additional research supplemented by qualitative interviewing may shed light on ratio-
nales for various tweeting behaviors and help assess whether journalists themselves
are aware of these gender imbalances. The gender asymmetry also begs future ques-
tions about gender-based public perception of journalists on social media. The
approach we have developed here has helped us advance our understanding of how
U.S. political journalists use Twitter, and how that use amplifies and reinforces gender
Usher et al. 17
imbalances. We hope other scholars in the future will adapt this approach to study the
same issue in other countries, advancing our understanding of cross-national differ-
ences and similarities in this area. We would hypothesize that variation in both institu-
tional structures (such as media systems) and in “political communication cultures”
(Pfetsch 2014) might influence outcomes in different countries.
Our exploratory approach matters at a conceptual level as well as a practical one,
especially in light of the recent attention to journalist-on-journalist sexual harassment.
As U.S. journalist Ezra Klein wrote in the aftermath of the “Shitty Media Men” list,
“What does it mean that these men—and so many others like them—held the power to
literally shape America’s political narrative?” (Klein, 2017). Of course, this study can-
not answer that question, but it does provide support to suggest that in addition to off-
line gender asymmetry, gender asymmetry in beltway journalism is also present
online. As a whole, our findings are deeply concerning: The gender imbalances pres-
ent on beltway journalism Twitter are another case showing women do not receive
adequate recognition or attention for their creative labor.
Beltway journalists are the embodiment of news as a political institution (Cook
2006). What happens in Washington matters on a global scale, and systemic gender
imbalances felt at the top of the political ecosystem may well have a trickle-down
effect, as framing models suggest (Entman 2004; Entman and Usher 2018). In
Washington politics, the framing of political communication happens by predominantly
male politicians and by beltway journalists who have troubling gender asymmetries,
including on Twitter. Our research calls for continuing to consider whose voices are not
heard (regardless of impact on coverage), and suggests that Twitter reflects (and may
amplify) existing gender asymmetry in the power center of U.S. politics. The signifi-
cance of the case deserves our alarm and possible attempts at mitigation.
Acknowledgments
Thank you to research assistants Scott Nover, Elise Hogue, and Jamie Reid.
Funding
The authors received no financial support for the research, authorship, and/or publication of this
article.
Notes
1. Complete data set, code, and output available on Github. Tool: https://gwu-libraries.github.
io/sfm-ui/. Notebook 1: https://nbviewer.jupyter.org/github/justinlittman/beltway_report-
ers/blob/master/jupyter/Genderdynamics-Part1.ipynb; Notebook 2: https://nbviewer.jupy-
ter.org/github/justinlittman/beltway_reporters/blob/master/jupyter/Genderdynamics-Part2.
ipynb.
18 The International Journal of Press/Politics 00(0)
ORCID iD
Justin Littman https://orcid.org/0000-0003-1527-0030
References
Adams, Catharine. 2017. “‘They Go for Gender First’: The Nature and Effect of Sexist Abuse
of Female Technology Journalists.” Journalism Practice. Published electronically August
9. doi:10.1080/17512786.2017.1350115.
Allsop, Jon, and Karen K. Ho. 2017. Global Media Confronts Its Own Weinsteins. Columbia
Journalism Review. https://www.cjr.org/watchdog/media-weinstein-sexual-harassment-
global.php.
Armstrong, Cory L. 2004. “The Influence of Reporter Gender on Source Selection in Newspaper
Stories.” Journalism & Mass Communciation Quarterly 81(1):139–154.
Armstrong, Cory L., & Gao, Fangfang. 2011. “Gender, Twitter and News Content: An
Examination Across Platforms and Coverage Areas.” Journalism Studies 12(4): 490-505.
Artwick, Claudette. 2014. “News Sourcing and Gender on Twitter.” Journalism 15 (8):
1111–27.
Bakshy, Eytan, Jake Hoffman, Winter Mason, and Duncan Watts. 2011. “Identifying Influencers
on Twitter.” Fourth ACM International Conference on Web Search and Data Mining, Hong
Kong 9–12 February.
Barnard, Stephen. 2014. “‘Tweet or Be Sacked’: Twitter and the New Elements of Journalistic
Practice.” Journalism 17 (2): 190–207.
Brems, Cara, Martina Temmerman, Todd Graham, and Marcel Broersma. 2017. “Personal
Branding on Twitter: How Employed and Freelance Journalists Stage Themselves on
Social Media.” Digital Journalism 5 (4): 443–59.
Carlson, Matt. 2017. Journalistic Authority: Legitimating News in the Digital Era. New York:
Columbia University Press.
Cha, Meeyoung, Hamed Haddadi, Fabrício Benevenuto, and Krishna P. Gummadi. 2010.
“Measuring User Influence in Twitter: The Million Follower Fallacy.” The Fourth
International AAI Conference on Weblogs and Social Media, Washington DC, USA 23-26
May.
Clayman, Steven, Marc Elliott, John Heritage, and Megan Beckett. 2012. “The President’s
Questioners: Consequential Attributes of the White House Press Corps.” The International
Journal of Press/Politics 17 (1): 100–21.
Congressional Directory. 2016. https://www.gpo.gov/fdsys/browse/collection.action?collectio
nCode=CDIR&browsePath=114%2F2016-02-12%2FR1&isCollapsed=false&leafLevelBr
owse=false&isDocumentResults=true&ycord=169
Cook, Timothy. 2006. “The News Media as a Political Institution: Looking Backward and
Looking Forward.” Political Communication 23 (2): 159–71.
Cozma, Raluca, and Kuan-Ju Chen. 2013. “What’s in a Tweet? Foreign Correspondents’ Use of
Social Media.” Journalism Practice 7 (1): 33–46.
Crouse, Timothy. (1973) 2013. The Boys on the Bus. New York: Random House.
Entman, Robert. 2004. Projections of Power: Framing News, Public Opinion, and US Foreign
Policy. Chicago: The University of Chicago Press.
Entman, Robert, and Nikki Usher. 2018. “Framing in a Fractured Democracy: Impacts of Digital
Technology on Ideology, Power, and Cascading Activation.” Journal of Communication 68
(2): 298–308.
Usher et al. 19
Frank, Russell. 2003. “These Crowded Circumstances: When Pack Journalists Bash Pack
Journalism.” Journalism 4 (4): 441–58.
Franks, Suzanne. 2013. Women and Journalism. London: I.B. Tauris.
Freelon, Deen, and David Karpf. 2015. “Of Big Birds and Bayonets: Hybrid Twitter Interactivity
in the 2012 Presidential Debates.” Information, Communication & Society 18 (4): 390–406.
Freelon, Deen, Charlton McIlwain, and Meredith D. Clark. 2016. “Beyond the Hashtags:
#Ferguson, #Blacklivesmatter, and the Online Struggle for Offline Justice.” Center for
Media and Social Impact, American University. http://cmsimpact.org/resource/beyond-
hashtags-ferguson-blacklivesmatter-online-struggle-offline-justice/.
Freelon, Deen, Charlton McIlwain, and Meredith D. Clark. 2018. “Quantifying the Power and
Consequences of Social Media Protest.” New Media & Society 20 (3): 990–1011.
George Washington University Libraries. 2016. “Social Feed Manager.” https://gwu-libraries.
github.io/sfm-ui/.
Global Media Monitoring Project. 2015. “GMMP 2015 Reports.” http://whomakesthenews.org/
gmmp-2015.
Gudipaty, Nagamallika. 2017. “Gendered Public Spaces: Online Trolling of Women Journalists
in India.” Comunicazione Politica 18 (2): 299–310.
Hamby, Peter. 2013. Did Twitter Kill the Boys on the Bus? Searching for a Better Way to Cover
a Campaign. Harvard: Joan Shorenstein Center on the Press, Politics, and Public Policy.
Hannis, Grant, and Cathy Strong. 2007. “The Visibility of Female Journalists in Australian and
New Zealand Newspapers: The Good News and the Bad News.” Australian Journalism
Review 29 (1): 115–25.
Hanusch, Folker. 2017. “Political Journalists’ Corporate and Personal Identities on Twitter
Profile Pages: A Comparative Analysis in Four Westminster Democracies.” New Media &
Society 20(4): 1488–1505. doi:10.1177/1461444817698479.
Hayes, Danny, and Jennifer L. Lawless. 2015. “A Non-Gendered Lens? Media, Voters, and
Female candidates in Contemporary Congressional Elections.” Perspectives on Politics 13
(1): 95–118.
Jürgens, Pascal, Andreas Jungherr, and Harald Schoen. 2011. “Small Worlds with a Difference:
New Gatekeepers and the Filtering of Political Information on Twitter.” Proceedings of
the 3rd International Web Science Conference. Koblenz, Germany 21, 15–17 June, 2011.
Klein, Ezra. 2017. “Political journalism has been profoundly shaped by men like Leon Wieseltier
and Mark Halperin.” Vox. https://www.vox.com/policy-and-politics/2017/10/27/16561386/
wieseltier-halperin-journalism-sexual-harassment.
Kreiss, Daniel. 2016. “Seizing the Moment: The Presidential Campaigns’ Use of Twitter During
the 2012 Electoral Cycle.” New Media & Society 18 (8): 1473–90.
Lasorsa, Dominic L. 2012. “Transparency and Other Journalistic Norms on Twitter.” Journalism
Studies 13 (3): 402–17.
Lasorsa, Dominic L., Seth C. Lewis, and Avery E. Holton. 2012. “Normalizing Twitter:
Journalism Practice in an Emerging Communication Space.” Journalism Studies 13 (1):
19–36.
Lawrence, Regina G. 2015. “Campaign News in the Time of Twitter.” In Controlling the
Message: New Media in American Political Campaigns, ed. Victoria A. Farrar-Myers and
Justin S. Vaughn, 93–112. New York: New York University Press.
Lawrence, Regina G., Logan Molyneux, Mark Coddington, and Avery Holton. 2014. “Tweeting
Conventions: Political Journalists’ Use of Twitter to Cover the 2012 Presidential Campaign.”
Journalism Studies 15 (6): 789–806.
20 The International Journal of Press/Politics 00(0)
Lucht, Tracy. 2016. “Job Satisfaction and Gender: Qualitative Differences at Iowa Newspapers.”
Journalism Practice 10 (3): 405–23.
Mantilla, Karla. 2013. “Gendertrolling: Misogyny Adapts to New Media.” Feminist Studies 39
(2): 563–70.
Matias, J. Nathan, Sarah Szalavitz, and Ethan Zuckerman. 2017. “FollowBias: Supporting
Behavior Change toward Gender Equality by Networked Gatekeepers on Social Media.”
Proceedings of the Conference on Computer Supported Cooperative Work and Social
Computing Portland, OR, USA 25 February-01 March 2017.
McGregor, Shannon C., Regina Lawrence, and Arielle Cardona. 2016. “Personalization, Gender,
and Social Media: Gubernatorial Candidates’ Social Media Strategies.” Information,
Communication & Society 20 (2): 264–83.
McGregor, Shannon C., and Rachel Mourão. 2016. “Talking Politics on Twitter: Gender, Elections,
and Social Networks.” Social Media & Society 2(3): 1–14. doi:10.1177/2056305116664218.
Meeks, Lindsey. 2013. “A Woman’s Place: Gender Politics and Twitter in the 2012 Elections.”
Doctoral dissertation, University of Washington, Seattle.
Megarry, Jessica. 2014. “Online Incivility or Sexual Harassment? Conceptualizing Women’s
Experiences in the Digital Age.” Women’s Studies International Forum 47 (A): 46–55.
Mills, Kay. 1997 “What Difference Do Women Journalists Make?” In Women, Media, and
Politics, ed. Pippa Norris, 41–55. New York: Oxford University Press.
Molyneux, Logan. 2015. “What Journalists Retweet: Opinion, Humor, and Brand Development
on Twitter.” Journalism 16 (7): 920–35.
Molyneux, Logan, and Rachel Mourão. 2017. “Political Journalists’ Normalization of Twitter:
Interaction and New Affordances.” Journalism Studies. Published electronically October 5.
doi:10.1080/1461670X.2017.1370978.
Mourão, Rachel. 2015. “The Boys on the Timeline: Political Journalists’ Use of Twitter for
Building Interpretive Communities.” Journalism 16 (8): 1107–23.
Mourão, Rachel, Trevor Diehl, and Krishnan Vasudevan. 2016. “I Love Big Bird: How
Journalists Tweeted Humor During the 2012 Presidential Debates.” Digital Journalism
4(2): 211-228.
Nee, Coates, and Judith Fusco Rebecca. 2015. “Tweets During Crisis Follow One-Way
Communication.” Newspaper Research Journal 36 (2): 197–211.
Nilsson, Monica Löfgren. 2010. “Thinkings” and “Doings” of Gender: Gendering Processes in
Swedish Television News Production.” Journalism Practices 4 (1): 1–16.
North, Louise. 2016a. “The Gender of ‘Soft’ and ‘Hard’ News: Female Journalists’ Views on
Gendered Story Allocations.” Journalism Studies 17 (3): 356–73.
North, Louise. 2016b. “Still a ‘Blokes Club’: The Motherhood Dilemma in Journalism.”
Journalism 17 (3): 315–30.
Nuernbergk, Christian. 2016. “Political Journalists’ Interaction Networks: The German Federal
Press Conference on Twitter.” Journalism Practice 10 (7): 868–79.
Olausson, Ulrika. 2017. “The Celebrified Journalist: Journalistic Self-Promotion and Branding
in Celebrity Constructions on Twitter.” Journalism Studies. Published electronically
August 2. doi:10.1080/1461670X.2017.1349548.
Parmelee, John H., Nataliya Roman, Berrin Beasley, and Stephynie C. Perkins. 2017. “Gender
and Generational Differences in Political Reporters’ Interactivity on Twitter.” Journalism
Studies. Published electronically September 1. doi:10.1080/1461670X.2017.1364140.
Patterson, Thomas E. 2017. “News Coverage of Donald Trump’s First 100 Days.” Cambridge:
Harvard University Shorenstein Center. https://shorensteincenter.org/news-coverage-don-
ald-trumps-first-100-days/.
Usher et al. 21
Pew Research Center. 2015. “Today’s Washington Press Corps, More Digital, Specialized.”
http://www.journalism.org/2015/12/03/todays-washington-press-corps-more-digital-spe-
cialized/.
Pfetsch, Barbara, ed. 2014. Political Communication Cultures in Europe: Attitudes of Political
Actors and Journalists in Nine Countries. New York: Palgrave Macmillan.
Poindexter, Paula. 2010. “Finding Women in the Newsroom and in the News.” In Women, Men
and News: Divided and Disconnected in the News Media Landscape, ed. Paula Poindexter,
Sharon Meraz, and Amy Schmitz Weiss, 33–46. New York: Routledge.
Reinardy, Scott. 2009. “Female Journalists More Likely to Leave Newspapers.” Newspaper
Research Journal 30 (3): 42–57.
Ritchie, Donald A. 2005. Reporting from Washington: The History of the Washington Press
Corps. New York: Oxford University Press.
Robertson, Nan. 1992. The Girls in the Balcony: Women, Men, and the New York Times. New
York: Random House.
Ross, Karen, Karen Boyle, Cynthia Carter, and Debbie Ging. 2018. “Women, Men and News:
It’s Life, Jim, But Not as We Know It.” Journalism Studies 19 (6): 824–45.
Steiner, Linda. 2012. “Failed Theories: Explaining Gender Difference in Journalism.” Review
of Communication 12 (3): 201–23.
Steiner, Linda. 2005. “The ‘Gender Matters’ Debate in Journalism: Lessons from the Front.” In
ed. Stuart Allan. Journalism: Critical Issues, 42–54, London: McGraw Hill.
Stevens, D., G. Kiger, and Patricia Riley. 2001. “Working Hard and Hardly Working: Domestic
Labor and Marital Satisfaction among Dual-Earner Couples.” Journal of Marriage and
Family 63 (2): 514–26.
Tsui, Celia, and Francis L. F. Lee. 2012. “Trajectories of Women Journalists Careers in Hong
Kong: A Field Theory Perspective.” Journalism Studies 13 (3): 370–85.
Van Zoonen, Liesbet. 2002. “One of the Girls? The Changing Gender of Journalism.” In News,
Gender and Power, ed. Stuart Allan, Gill Branston, and Cynthia Garter, 33-46. New York:
Routledge.
Volz, Yong Z., and Francis L. F. Lee. 2013. “What Does It Take for Women Journalists to Gain
Professional Recognition? Gender Disparities among Pulitzer Prize Winners, 1917–2010.”
Journalism & Mass Communication Quarterly 90 (2): 248–66.
Women’s Media Center. 2017. “The Status of Women in the US Media: 2017.” http://www.
womensmediacenter.com/assets/site/reports/10c550d19ef9f3688f_mlbres2jd.pdf.
Zelizer, Barbie. 1993. “Journalists as Interpretive Communities.” Critical Studies in Media
Communication 10 (3): 219–37.
Author Biographies
Nikki Usher, PhD, is associate professor at the School of Media and Public Affairs at George
Washington University. She is the author of two books, Making News at the New York Times,
and Interactive Journalism: Hackers, Data, and Code.
Jesse Holcomb, MA, is assistant professor of Journalism and Communication at Calvin College
in Grand Rapids, Michigan. He is also a fellow at Columbia University’s Tow Center for Digital
Journalism.
Justin Littman is librarian/software developer at George Washington University Libraries. As
a member of the team that developed the Social Feed Manager tool, his work involves support-
ing social media research and archiving.