A comparative analysis of source and reporter gender in newsrooms managed
by men and women
By Stephanie Craft
School of Journalism
University of Missouri
Columbia, MO 65211-1200
[log in to unmask]
School of Journalism
University of Missouri
Columbia, MO 65211-1200
[log in to unmask]
and Cheolhan Lee
School of Journalism
University of Missouri
Columbia, MO 65211-1200
[log in to unmask]
**Manuscript submitted to the Newspaper Division for consideration of
presentation at the annual convention of the Association for Education in
Journalism and Mass Communication, Kansas City, July/August, 2003
**Craft is an assistant professor, Wanta a professor and Lee a doctoral
student in the Missouri School of Journalism, Columbia, MO 65211-1200
Efforts to increase the diversity of newsroom staffs in terms of gender,
race and ethnicity rest in part on a kind of conventional wisdom that these
demographic differences can (and even ought) to produce diversity in the
types of news covered or in the ways news is covered. But do those
characteristics really matter when it comes to shaping the news? Research
seems to suggest that the answer is both yes and no.
Gatekeeping theory, initially applied to news in David Manning White's
study of a wire editor he dubbed "Mr. Gates," suggests that individual
journalists shape the news through their decisions about which potential
news items do and do not pass through the gates and into the newspaper or
newscast. A gatekeeper's personal beliefs, White found, influenced those
decisions. A broad range of research, however, suggests that these
individual influences on the news are muted by news routines and
conventions (e.g., Shoemaker, Eicholz, Kim and Wrigley, 2001; Shoemaker and
Reese, 1996; Berkowitz and Beach, 1993; Brown, Bybee, Weardon and
Straughan, 1987). More general models of news production acknowledge the
importance of external factors in shaping the news. McQuail's model (1993),
for example, places the media organization, not the individual journalist,
at the center of a web of social forces that include sources, advertisers,
competitors, and the like. The individual within the organization is
subject to all the pressures impinging on the organization.
Research also has addressed the factors that help shape those news routines
and conventions. For example, researchers have attempted to discover what
impact civic journalism, which calls for setting aside many traditional
news routines, has on the news (Kurpius, 2002; Massey, 1999). Others have
examined the largely "male" culture of the newsroom (e.g., Turk, 1987),
noting its persistence despite the increasing number of women in the
The present study seeks to add to the literature on the impact of
individual characteristics on the news, in particular addressing whether
one can link gender to differences in practices related to use of sources.
More specifically, do journalists in a newsroom with a relatively high
percentage of female gatekeepers differ in their use of sources compared
with journalists in more male-dominated newsrooms?
Women in the newsroom
Weaver and Wilhoit's portrait of journalists shows that the typical
journalist is male, Protestant, liberal, college-educated and middle-class
(Weaver & Wilhoit, 1996). That these numbers do not represent the
demographics of the American public generally is one reason, among others,
that a number of journalism organizations, including the American Society
of Newspaper Editors, have made increasing newsroom diversity a top
priority. In addition, a broad range of research had demonstrated that
women and minorities are often portrayed in stereotypical ways or are
covered differently from men in the news (see, for example, Kahn, 1994 and
Devitt, 2002, which address the treatment of female political candidates in
Though women journalists began making inroads into what had been a largely
male occupation long ago, it is only in the past 20 years that they have
ascended to higher levels of management. Even so, their representation at
the top does not match their representation in the profession overall, much
less in society. A recent census of newsroom employment found that women
constituted 37 percent of the newsroom workforce; and 34 percent of all
newsroom supervisors are women (ASNE, 2001). A 2002 study of women in
newspapers found "only 18 percent of the corporate executives at assistant
vice president or higher are women" (Hemlinger and Linton, 2002, p. 16).
Cultural barriers are often cited as the explanation of such findings.
Indeed research (e.g., Cann and Mohr, 2001; Turk, 1987; Brown, Bybee,
Weardon & Straughan, 1987) suggests that male culture in the newsroom may
explain both the relative lack of women in higher posts as well as the
continuation of norms and routines that serve to standardize the news and
diminish the potential for individual characteristics to have an impact on
it. Burks and Stone (1993) found few differences in the professional
experiences, including management styles, of female and male television
news directors, pointing to a consistency of newsroom operations,
regardless of managers' gender.
As van Zoonen (1988) and Liebler (1994) point out, to argue that the
presence of greater numbers of women in the newsroom can have an impact on
the news itself assumes an autonomy for individual journalists that is not
borne out by research on journalistic routines and conventions. Moreover,
Liebler's study indicated no difference in the sense of personal autonomy
reported by male and female journalists, regardless of the numbers of women
in the newsroom.
Bleske replicated White's gatekeeping study, this time with a female wire
editor, and found no effects that could be attributed to gender (Bleske,
1991). However, there is evidence that male and female reporters are
routinely assigned stories on the basis of gender. Craft and Wanta (2002),
for example, found male and female reporters covered different issues men
covering traditionally "hard" news and women more features and "soft" news
but only in newsrooms with predominantly male editors. That is, in
newsrooms with female managers, reporter assignments did not appear to
differ according to reporter gender. So, if women are relegated to "women's
news," the extent to which they can bring a "woman's perspective" to other
news is limited (Singleton and Cook, 1982). A content analysis of 20 years
of women's magazines indicated that the presence of female editors did not
reduce stereotypical portrayals of women.
Sources and the news
Gans (1979) considered sources to be of primary significance in explaining
the news. Sources can affect the content of news, through psychological
pressures on the reporter to maintain relationships (Manoff and Schudson,
1986) and avoid angering sources. Social location and organizational
routine affect source selection by restricting "how wide journalists cast
their nets for news" and, therefore, the kinds of news that is covered
(Manoff and Schudson, 1986, p. 16). Sigal (1973) found that Washington Post
and New York Times reporters relied on "routine channels" of information to
report stories, a finding reconfirmed by later research (Brown, Bybee,
Wearden and Straughan, 1987). These studies and others document
journalists' reliance on elite, government and largely male news sources.
For example, Hallin, Manoff and Weddle (1993) confirmed previous research
indicating reporters' reliance on elite news sources. They also found that
97 percent of the sources national security reporters used were male.
It is important to note, as Voakes, Kapfer, Kurpius, and Chern (1996)
argue, that source diversity and content diversity are distinct and vary
independently. Their study did not address gender diversity specifically,
but their definition of source diversity as a distribution of sources
across status, affiliation and other variables can be expanded to include
gender, Kurpius (2002) argues.
Source and reporter gender
Researchers have found significant differences between male and female
newspaper reporters in their selection of male and female sources and
prominence given to those sources in news stories (Armstrong, 2002; Zoch &
Turk, 1999; and Liebler & Smith, 1997). In broadcast, Smith and Wright
(1998) found no significant link between proportion of women in local
television news management and use of female sources. Can and Mohr (2001)
also found no differences in source selection by male and female reporters
on primetime news broadcasts in Australia, but significant differences in
the length of stories reported by male and female reporters and the topics
they covered. Results of a study by Grabe, Zhou and Barnett (1999),
however, suggested that female reporters seek out female sources.
Massey (1999) found that women appeared as information sources more often
under civic journalism, but only marginally and only in stories reported by
female journalists. In another civic journalism study by Kurpius (2002),
female civic journalists included more female sources in their stories than
did male civic journalists, though the use of female sources overall did
not reflect U.S. Census figures.
Previous research both suggests that source selection may differ according
to reporter gender and that it is difficult to establish those differences
given journalistic routines that mute individual impact on the news and
favor certain kinds of sources. The present study, then, seeks to examine
whether source selection is related to a predominance of women in positions
of power in the newsroom and not individual journalist gender alone. The
cited literature suggests the following hypotheses:
H1: The use of male and female sources will differ according to whether a
newspaper's editors are predominantly men or women.
H2: Female reporters will use a greater percentage of female sources than
male reporters use.
H3: The source selection strategies of male and female reporters will be
influenced by the predominant gender of their editors.
Articles from 30 U.S. daily newspapers, selected at random from Editor and
Publisher Yearbook, were analyzed in this study. Using random number tables
to determine starting pages and subsequent listings, the researchers
compiled a list of newspapers, noting the number of job positions listed
under the headings "News Executives" and "Editors" and the number of those
positions held by men and women. Names that are not unquestionably "male"
or "female" were counted separately.
Where possible, researchers sought to balance the geographic distribution
of papers in each category and the overall sample. Newspapers that did not
have websites were eliminated from the sample. Table 1 lists the newspapers
that were included in the study.
Articles for the analysis were selected from a constructed week beginning
February 21, 2002, and ending March 9, 2002. All local news stories
defined as those stories with local staff bylines were included. Because
many of the smaller newspapers do not publish Sunday editions, Sunday was
eliminated from the week. In cases when stories were not available, stories
from the next available day were used. For instance, three newspapers did
not publish on Saturday, so the following Monday was included. One
newspaper did not publish on Saturday or Monday. In this case, we included
the following Tuesday and Wednesday.
It should be noted that the stories included in the analysis were only
those that the newspapers had uploaded to their websites. While this
method did not necessarily mean that all local stories published on any
given day were included in our analysis, it did provide us with a good
range of stories, and certainly included the most important stories of the
day. The total number of stories analyzed was 818.
The unit of analysis for this study was the newspaper article. Three coders
conducted the analysis. Coders recorded the newspaper circulation size and
the percentage of management positions held by women. Coders also noted
reporter gender based upon the article byline. Reporter gender for articles
without bylines and articles with bylines that could be either "male" or
"female" was not coded.
Sources in stories were coded both for gender and for status. Following a
scheme developed by Cann and Mohr (2001), status was determined by
examining the source's role within the context of the story. "[W]here the
occupation of the source was given and appeared relevant to the story, the
source was coded as expert. If the source was interviewed as a bystander,
witness, relative, or victim, or otherwise signified a general public
response … the source was coded as non-expert" (Cann & Mohr, 2001, p. 167).
Therefore, sources were given one of the following codes: male expert, male
non-expert, female expert, or female non-expert. Where the gender of the
source could not be determined, it was not included in the analysis.
A total of 1,766 sources were included in the analysis. Of these, 1,102,
or 62.4 percent, were male experts, 333 (18.9 percent) were female experts,
178 (10.1 percent) were male non-experts and 153 (8.7 percent) were female
non-experts. Clearly, male experts were highly relied upon by the
reporters in the study.
Table 2 lists the frequencies and the chi-square results for the use of
sources across the 30 newspapers in our study. Male experts were
predominantly used at both newspapers with high and low percentages of
women in managerial positions, but especially so at high women
newspapers. Interestingly, female experts were more likely to be used at
the low women newspapers. Female and male non-experts were used similarly
across newspapers. The difference in source use across the newspapers was
statistically significant (chi-square = 17.43, p = .017). These findings
contradict the expectation of Hypothesis 1.
Table 3 shows the results of the analysis examining source use by male and
female reporters. The differences here also were statistically significant
(chi-square = 8.828, p = .032). Female reporters clearly were more likely
to use female experts and non-experts than were male reporters. Male
reporters, on the other hand, used more male experts and non-experts.
Therefore, Hypothesis 2 was supported.
The results examining differences for male and female reporters across
newspapers with high and low percentages of women in managerial positions
appear in Table 4. Male and female reporters did not differ in their use
of sources at newspapers with a high percentage of women in managerial
positions. However, at male dominated newspapers, male and female
reporters did differ. Here, male reporters were more likely to use male
non-experts than were female reporters. Female reporters were more likely
to use female non-experts than were male reporters. Reporter gender did
not matter in the use of expert sources, however. Both male and female
reporters used male experts more than 50 percent of the time and female
experts about 23 percent of the time. These findings support Hypothesis 3.
The number of sources per story did not differ significantly across any of
the variables here. Newspapers with a high percentage of women managers
averaged 2.57 sources per story, and newspapers with a low percentage of
women managers averaged 2.36 sources. Male reporters used 2.82 sources per
story, and female reporters used 2.68.
The present study examined gender differences in the use of sources in
news stories. The study had two main concerns here: 1. whether the gender
of reporters mattered in the type of sources selected for use in stories;
2. whether newspapers with high percentages of women in management differed
from newspapers with low percentages of women in management in their use of
sources. The results point to several differences.
First, journalists at male-dominated newspapers actually used more female
experts and fewer male experts in their news stories. This is the opposite
of what might be expected. If male editors are predominant in a
newspaper's management structure, reporters logically should be more likely
to use more traditional sources, concentrating on male experts. Our
content analysis, however, found the opposite. The reasons for this
finding are puzzling.
Possibly, women in management may feel the need to prove their abilities
within the traditionally male-dominated business and thus encourage their
reporters to gather information from traditional sources such as male
experts. Men in male-dominated newsrooms, on the other hand, may feel the
need to prove that they are not biased and therefore attempt to use more
female sources in reporting. Certainly, this bears further research.
In all 30 newspapers, female reporters, meanwhile, used more female
sources both experts and non-experts in their stories. This, of course,
could have been because of a conscious effort on the part of female
reporters to get women's perspectives in their stories. It also could have
been because female reporters feel more comfortable with interviewing women
for their stories than do male reporters.
On the other hand, female reporters may have been placed in beats in which
female sources are more important the education beat or the lifestyle
section, for example. Male reporters, meanwhile, may still dominate in the
more traditional beats, such as politics, as Craft and Wanta (2002) found.
Another finding also points to this final explanation: Female and male
reporters at male-dominated newspapers differed in their use of
sources. Here, female reporters were more likely to use female non-experts
and male reporters were more likely to use male non-experts. The gender
difference was found for non-expert sources only. Again, beat differences
may have been the cause here. If male and female reporters are placed in
traditionally gender-based beats men in hard news beats and women in soft
news, feature beats they would be more likely to use sources that match
the gender of their beats male non-experts for male reporters and female
non-experts for female reporters.
Male and female reporters at newspapers with a high percentage of women in
management, on the other hand, did not differ in their use of
sources. Male experts dominated in all stories, regardless of the gender
of the reporter. This again may show that, in newspapers in which women
hold important managerial positions, female reporters are on equal footing
with male reporters. Since men and women are not relegated to traditional
gender-based beats, they tend to report similar stories and thus tend to
use similar sources. Thus, from a socialization of the news perspective,
male and female reporters conduct their news routines in a similar manner
if the management of their newspaper puts them on equal terms.
Another indication of support for a socialization of news perspective can
be seen in the frequency of sources used by reporters. Male and female
reporters used a similar number of sources per story. Similarly,
newspapers with a larger percentage of women in management did not differ
from newspapers with a small percentage of women in management on the
number of sources used. Thus, news stories were constructed similarly in
many of the comparisons here.
Overall, the findings of this study demonstrate a tendency for male and
female reporters to use sources based on their gender. Moreover, the gender
makeup of the newsroom appears to have some influence on source selection
practices as well, though not, perhaps, in the ways we might anticipate.
Certainly further research needs to address how the culture of the newsroom
contributes to these differences.
Table 1. Newspapers included in the study, by relative percentage of
High percentage Percent of
The (Middlesboro, KY) Daily News 72.7
Americus (GA) Times-Recorder 83.3
(Ionia, MI) Sentinel-Standard 100
New Braunfels (TX) Herald-Zeitung 100
Baraboo (WI) News Republic 75.0
The (Hanover, PA) Evening Sun 54.5
Cleveland (TN) Daily Banner 81.8
(Eureka, CA) Times-Standard 62.5
Martinsville (VA) Bulletin 62.5
The (Maryville, TN) Daily Times 50.0
Pensacola (FL) News-Journal 52.4
The (Quincy, MA) Patriot Ledger 54.1
The Knoxville (TN) News Sentinel 50.5
The Trentonian 66.7
Kalamazoo Gazette 50.0
(Mount Carmel, IL) Daily Republican Register 0
Los Alamos (NM) Monitor 42.9
The Ashland (OR) Daily Tidings 8.3
The Logan (WV) Banner 0
Edwardsville (IL) Intelligencer 0
(Redding, CA) Record Searchlight 28.6
Lancaster (OH) Eagle-Gazette 38.5
The (Norristown, PA) Times Herald 42.9
(Lafayette, IN) Journal and Courier 30.0
Northwest Florida Daily News 25.0
The Vancouver (WA) Columbian 12.5
(Youngstown, OH) Vindicator 35.5
The (Bergen County-Hackensack, NJ) Record 31.4
The (Springfield, MA) Union-News 21.2
The (Santa Rosa, CA) Press Democrat 25
Table 2. Source use at newspapers with high and low percentages of women
in managerial positions (frequencies and column percentages).
High women Low women
Male expert 622 (67.1%) 623 (58.2%)
Female expert 148 (16.0%) 230 (21.5%)
Male non-expert 85 (9.2%) 122 (11.4%)
Female non-expert 72 (7.8%) 95 (8.9%)
Chi-square = 17.43 (3 d.f.), p = .001
Table 3. Source use by male and female reporters (frequencies and column
Female reporters Male reporters
Male expert 397 (59.5%) 705 (64.1%)
Female expert 143 (21.4%) 190 (17.3%)
Male non-expert 60 (9.0%) 118 (10.7%)
Female non-expert 67 (10.0%) 86 (7.8%)
Chi-square = 8.828 (3 d.f.), p = .032
Table 4. Source use by male and female reporters at newspapers with high
and low percentages of women in managerial positions (frequencies and
High women newspapers
Female reporters Male reporters
Male expert 172 (68.3%) 385 (67.3%)
Female expert 42 (16.7%) 78 (13.6%)
Male non-expert 20 (7.9%) 56 (9.7%)
Female non-expert 18 (7.1%) 53 (9.3%)
Chi-square = 2.693 (3 d.f.), p = .441
Low women newspapers
Female reporters Male reporters
Male expert 225 (55.6%) 287 (58.3%)
Female expert 91 (22.5%) 112 (22.8%)
Male non-expert 40 (9.9%) 62 (12.6%)
Female non-expert 49 (12.1%) 31 (6.3%)
Chi-square = 10.131 (3 d.f.), p = .017
Armstrong, C.L. (2002). The influence of reporter gender on source
selection in newspaper stories. Paper presented at the annual meeting of
the Association for Education in Journalism and Mass Communication, Miami, FL.
American Society of Newspaper Editors, 2001. Newsroom employment census
[online]. Washington, DC: Author. Available from World Wide Web:
Berkowitz, D., & Beach, D. W. (1993). News sources and news context: The
effect of routine news conflict and proximity. Journalism Quarterly, 70, 4-12.
Brown, J. D., Bybee, C. B., Wearden, S. T., & Straughan, D. M. (1987).
Invisible power: Newspaper news sources and the limits of diversity.
Journalism Quarterly, 64(1), 45-54.
Burks, K.K., and Stone, V. A. (1993). Career-related characteristics of
male and female news directors. Journalism Quarterly. 70 (3, Autumn), 542-549.
Cann, D.J. & Mohr, P.B. (2001). Journalist and source gender in Australian
television news. Journal of Broadcasting & Electronic Media, 45 (1), 162-174.
Craft, S. & Wanta, W. (2002). Women in the newsroom: Influences of female
editors and reporters on the news agenda. Paper presented at the annual
meeting of the Association for Education in Journalism and Mass
Communication, Miami, FL.
Devitt, J. (2002). Framing gender on the campaign trail: Female
gubernatorial candidates and the press. Journalism & Mass Communication
Quarterly, 79 (2), 445-463.
Grabe, M.E., Zhou, S., and Barnett, B. (1999). Sourcing and reporting in
news magazine programs: 60 Minutes versus Hard Copy. Journalism & Mass
Communication Quarterly. 76 (2, Summer), 293-311.
Hemlinger, M.A. & Lindon, C.C. (2002). Women in newspapers: Still fighting
an uphill battle. Evanston, IL: Northwestern University, Media Management
Jolliffe, L. & Catlett, T. (1994). Women editors at the 'Seven Sisters'
magazines, 1965-1985: Did they make a difference? Journalism Quarterly 71,
Kahn, K. (1994). The distorted mirror: Press coverage of women candidates
for statewide office. Journal of Politics, 56, 154-173.
Kurpius, D.D. (2002). Sources and civic journalism: Changing patterns of
reporting? Journalism & Mass Communication Quarterly, 79 (4), 853-866.
Liebler, C. (1994). How race and gender affect journalists' autonomy.
Newspaper Research Journal, 15 (3), 122-130.
Liebler, C.M. & Smith, S.J. (1997). Tracking gender differences: A
comparative analysis of network correspondents and their sources. Journal
of Broadcasting & Electronic Media, 41 (1), 58-68.
Massey, B.L. (1999). Civic journalism and gender diversity in news-story
sourcing. Paper presented at the annual meeting of the Association for
Education in Journalism and Mass Communication, New Orleans, LA.
Powers, A., & Fico, F. (1994). Influences on use of sources at large U.S.
newspapers. Newspaper Research Journal, 15(4), 87-97.
Shoemaker, P.J., Eichholz, M., Kim, E., and Wrigley, B. (2001). Individual
and routine forces in gatekeeping. Journalism & Mass Communication
Quarterly, 78 (2), 233-246.
Shoemaker, P.J. & Reese, S.D. (1996). Mediating the message: Theories of
influences on mass media content. New York: Longman.
Singleton, L.A. & Cook, S.L. (1982). Television network news reporting by
female correspondents: An update. Journal of Broadcasting 26, 487-491.
Smith, L.K. & Wright, J.W. II (1998). Women in television news management:
Do they make a difference? Paper presented at the annual meeting of the
Association for Education in Journalism and Mass Communication, Baltimore, MD.
Voakes, P.S., Kapfer, J., Kurpius, D., & Chern, D.S. (1996). Diversity in
the news: A conceptual and methodological framework. Journalism and Mass
Communication Quarterly, 73(3), 582-593.
Weaver. D. & Wilhoit, G. (1996). The American Journalist in the 1990s: I.S.
News People at the End of an Era. Mahwah, NJ: Lawrence Erlbaum Associates.
Ziegler, D. & White, A. (1990). Women and minorities on network television
news: An examination of correspondents and newsmakers. Journal of
Broadcasting & Electronic Media, 34(2), 215-223.
Zoch, L.M. & Turk, J.V. Women making news: Gender as a variable in source
selection and use. Journalism & Mass Communication Quarterly 75(4), 762-775.
van Zoonen, L. (1988). Rethinking women and the news. European Journal of
Communication. 3 (1, March), 35-53.