Home > coverstory > The Gender Gap:Women Are Still Missing as Sources for Journalists

Despite rising numbers of women in the workforce and in journalism schools, the news of the day still largely comes from a male perspective, according to a new study of press coverage.

A broad look across the American news media over the course of nine months reveals that men are relied on as sources in the news more than twice as often as women, a study by the Project for Excellence in Journalism has found.

More than three quarters of all stories contain male sources, while only a third of stories contain even a single female source, according to the study, which was drawn from an examination of 16,800 news stories across 45 different news outlets during 20 randomly selected days over nine months.

The disparity, moreover, holds true across newspapers, cable, network news and the online world.

The findings may strike some observers as ironic given the efforts of many news outlets to increase their audience by reaching out to women梐nd particularly to younger women, a group that generally is under represented as news consumers.

Among the findings:

In every topic category, the majority of stories cited at least one male source.
In contrast, the only topic category where women crossed the 50% threshold was lifestyle stories.
The subject women were least likely to be cited on was foreign affairs.
Newspapers were the most likely of the media studied to cite at least one female source in a story (41% of stories). Cable news, despite all the time it has to fill, was the least likely medium to cite a female source (19% of stories), and this held true across all three major cable channels.
On network TV, the morning news programs, which often cover lighter fare, relied more on female sources. The evening newscasts were somewhat less likely, but still did so more than cable.
The sports section of the newspaper stood out in particular as a male bastion. A mere 14% of stories on the front page of the sports section cited a woman, versus 86% that contained at least one male source.
The study by the Project, a research institute affiliated with the Columbia University Graduate School of Journalism and funded by the Pew Charitable Trusts, examined 16 newspapers from a range of circulation categories, four nightly newscasts (commercial networks and PBS), three network morning news shows, nine different cable programs, and nine Web sites studied at four different times during the day. The study counted all sources whose gender could be determined by their appearance, a typically female name, self-identification, or some other form of positive identification. A source is defined as anyone providing information to the report, be it through a direct quotation, indirect quotation, or as the person to whom data or other information is attributed.
The numbers suggest that the representation of women as sources in the news has a significant distance to go towards reflecting their role in American society generally. Women account for 52% of the country抯 population and roughly 47% of the employed civilian workforce, according to 2000 data from the U.S. Census. What抯 more, their presence in management positions is not far behind�% of those working in management, business and financial operations are women.

Their representation in politics is relatively strong as well. For instance, a full 44% of press secretaries in the U.S. House of Representatives, the staffers to whom reporters often speak, are women, and they account for 51% of House staffers overall.(1) On the other hand, when it comes to elected officials women are still behind. They hold only four of 21 presidential cabinet seats and only 69 seats in the House of Representatives out of 440.

In the news, however, the numbers come closer to reflecting those of elected officials than the American workforce. Looking across all media, three-quarters (76%) of the stories studied contained at least one male source. Just a third (33%) contained a female source.

The gap between the genders grows even larger if we raise the bar to two or more sources. Reporters were more than three times as likely to cite two or more males within a news story as to cite at least two females (55% versus 14%). This suggests that the orientation toward males goes beyond the primary source in a story. Finding a male as the best first source does not apparently lead a journalist to look for a female as the second or third source.

Male and Female Sources in the News
Percent of all Stories
Males Females
0 24% 67%
1 21 20
2+ 55 14
Totals may not equal 100 due to rounding

There is no suggestion here that journalists should seek gender balance in every story. Certainly in some stories the most appropriate sources might be male just as in others they might be female. Instead, the study was designed to look across a wide spectrum of news coverage and media to get a basic idea of gender representation in the news.

The dominance of male sources over female exists in all media, though in some more heavily than in others.

Female Versus Male Sources, Percent of all Stories
Newspapers Online Network Evening PBS NewsHour Network Morning Cable
Female 1+ 41% 36% 27% 17% 34% 19%
Male 1+ 88% 89% 63% 59% 55% 53%
Not only were newspapers more than twice as likely as cable news to cite even one female source, they were also more likely than other media to cite two or more.

The study examined all news stories found on page A1, the front page of the metro section and the front page of the sports section for 16 different newspapers across four circulation size categories�589 stories in all. (1)

Overall, 41% of print stories contained at least one female source, and 19% reached the higher threshold of citing two or more. One such story ran in the Philadelphia Inquirer on August 10, 2004. The A1 piece about remarks Mrs. Laura Bush made about stem-cell research quoted both Mrs. Bush and Mary Rachel Faris, a hematologist at Abington Memorial Hospital in Pennsylvania.

Nevertheless, print stories were still half as likely to contain a female source as a male source (88% cited at least one male source). Even the stem-cell story in the Inquirer cited four male sources.

The type of newspaper story also made a difference. Wire service stories were less likely to cite females than were reports written by the newspaper抯 own staff members. Staff-written stories were about twice as likely as wire service stories to contain a female source (47% versus 25%). Stories that were a combination of staff and wire copy fell in between (37%).

It is not simply that wire stories use fewer sources. The same kind of gap did not occur with male sources. Here staff written pieces and wire copy were roughly equal. A full 91% of staff written stories cited at least one male as did 87% of wire copy and 89% stories that combined staff and wire material.

Gender of Sources in Newspaper Stories, Staff Versus Wire Copy
Staff Wire Combo
Female Male Female Male Female Male
None 53% 9% 75% 13% 63% 11%
1 or more 47 91 25 87 37 89
Totals may not equal 100 due to rounding.

The size or circulation of a newspaper also seemed to make a difference. Bigger papers included more female voices (46% of stories in the largest papers versus 43% in midsize and 33% in the smallest). Smaller papers also tended to carry more wire service reports than did large papers, perhaps accounting for some of the disparity.

Another difference emerged in the various sections examined: Page A1, the metro section-front and the sports section-front.

The front of the metro section was the most inclusive of women梒iting them in more than half of its coverage, 57% of all stories. Page A1 was slightly behind at 50%. But on the front-page of the sports section, a mere 14% of stories included a female voice. This was the mirror image of male sources, who were cited in 86% of stories and not cited in 14%. Even in the era of Title IX and the push for women抯 sports, the sports section-front stands out for its lack of female voices.

(A separate study by Terry Adams and C.A. Tuggle suggests similar disparities may exist on cable sports programs. Their 2002 study found that women were the subjects of less than 5% of coverage on ESPN抯 SportsCenter. (2))

Has the 24-hour news culture been more inclusive of women? Hardly. Females fared worse here compared with any other medium studied. Of the roughly 6,550 cable stories examined on CNN, FOX and MSNBC, just 19% cited a female source.

The 20-day study spread across nine months of the year included three program types from each network representative of three distinct parts of the cable day: daytime programming, the closest program to a traditional newscast and the highest-rated prime time talk show on each channel. These criteria resulted in the following programs: The 11-12 o抍lock hour at each network; CNN抯 揘ewsnight with Aaron Brown� and 揕arry King Live� FOX抯 揝pecial Report with Brit Hume� and 揟he O扲eilly Factor� MSNBC抯 揅ountdown with Keith Olbermann� and 揌ardball with Chris Matthews.�br>
Past research has found that cable news stories tend to cite fewer sources overall than do other news media. (1) This was borne out in the examination of gender as well. Indeed, cable stories were less likely than other mediums to even cite a male source (53% on cable versus 63% on network evening news and 88% in newspapers).

Nevertheless, the gap between male and female sources stood out. While most mediums were roughly twice as likely to cite a male as a female source, cable stories were nearly three times as likely (19% female versus 53% male.) The gap was so great on cable that these stories were even more likely to have two or more male sources (21%) than to have just one or more female sources.

On three different nights of MSNBC扴 interview-style program 揌ardball with Chris Matthews� (April 15, May 4 and June 16, 2004), every single guest on the show was a man. The segments ranged from the Iraqi prison scandal discussed with members of Congress, journalists covering the story and former military personnel, to a discussion with Kareem Abdul-Jabbar about his new book, to the role of religion in politics, to election advertising strategies of the Bush and Kerry.

In one segment, Matthews ran a video clip of the wife of a solider accused in the prison scandal. Otherwise, every single source on these three nights梘uest or other梬as male.

Cable Channel Similarities

Looking more closely, the three cable channels appear quite similar when it comes to the gender of sources. Females were cited in just 15% of stories studied on MSNBC, 21% on CNN and 19% on Fox. Conversely, at least 50% of the stories on each of the three networks cited a male source.

Female Versus Male Sources on the Three Cable News Channels, percent of all Stories
Females 1+ 21% 19% 15% 19%
Males 1+ 50% 61% 51% 53%

Among the three program types examined, the prime time talk shows were the most inclusive of women (especially 揕arry King Live�). Among these talk shows, 28% of the segments cited at least one female source.

The 11 A.M. live news hour was the least inclusive program type. Just 11% of stories during this daytime period cited a female source條ess than a third of those that cited a man (37%).

The evening cable newscasts had more sourcing, but the ratio of female versus male sources was not much different. Stories on these programs were also about three times as likely to cite a male as a female (58% male versus 19% female) and were even more likely to cite two or more male sources than to cite a single female source (34% contained two or more male sources).

(1) The Project for Excellence in Journalism, The 2005 Annual Report on the State of the News Media, Cable content chapter, http://www.stateofthemedia.org/2005/narrative_cabletv_contentanalysis.asp?cat=2&media=5 , 2005.
Across media, are women more likely to be cited in some story topics than in others? To a certain degree the answer is yes, but still at levels much behind that of men. In fact, the only topic area where women were cited in more than half of the stories was lifestyle stories such as one on the CBS 揈arly Show� about what to do if someone sees a child getting into a car with an intoxicated adult. Beyond lifestyle, the other topics most inclusive of women sources were government (44%), crime (39%) celebrity/entertainment (38%) and accidents (38%).

Even among these, though, male voices still dominate. In all five topic categories, at least one male is cited in more than half of the stories. Celebrity/entertainment stories were the least likely to use a male source but still, 56% cited them. (The only category with a lower percentage (45%) is that of miscellaneous stories which do not fit into one of the main topic areas.

Where are male voices most likely to emerge? In the more traditional categories of government (86%), campaigns and elections (84%), foreign affairs (80%) and defense and military (76%).

Female and Male Sources for Different Story Topics, Precent of all Stories
Female Sources Male Sources
0 1+ 0 1+
Government 57% 44% 14% 86%
Defense/Military 78 22 24 76
Foreign Affairs 80 20 20 80
Elections 66 34 16 84
Domestic Affairs 63 37 27 73
Crime 61 39 29 71
Business 67 33 40 60
Celebrity/Entertainment 62 38 44 56
Lifestyle 47 53 32 68
Science/Technology 66 34 26 74
Accidents 62 38 35 65
Other 71 29 56 44
Totals may not equal 100 due to rounding

Differences Among Media

Newspapers� relative strength in including female voices carries through across most topics. At least one female voice was cited in the majority of stories studied in seven out of the twelve topic categories in print. Across all media combined, only one topic category cited a woman in at least 50% of stories梐nd that was lifestyle.

In print, lifestyle was again the topic most likely to cite women (66%), followed by domestic affairs (61%), election stories (57%), and government stories (56%). The greatest divergence from other media came in election coverage. Across all media, it came in seventh in its use of female sources (tied with science and technology at 34% of stories) versus third within newspapers.

Even in some of these stories, though, the female voice was far from dominant. A New York Times story from March 19 reported on Senator McCain抯 comments about John Kerry抯 defense record. Times journalist Todd S. Purdum first quoted Mr. McCain, Bush spokesman Terry Holt, and DNC spokesman Jano Cabrera. It was not until the last three paragraphs that a female voice was brought in, that of Victoria Clarke, a former press secretary for both McCain and later the Pentagon.

Online news coverage was similar to print in its overall reliance on women sources, but differed in the tendencies within story topics. Five topic categories cited a woman in at least half the stories, but they differed from the newspaper topics. The 揗iscellaneous� grouping梥tories which have little connection to other current events, was the most common (68%). This was followed by lifestyle stories (61%) such as one on CNN.com about snowboarding that included remarks from snowboarder Tricia Byrnes. Next came accidents (57%), celebrity (57%) and business (51%).

On commercial network evening news, just three topic categories included a female voice in half or more of the stories. At the top were celebrity/entertainment stories (58%) and lifestyle stories (58%), though these are also among the least common story topics on commercial evening news, (accounting for just 7% of the coverage studied). On February 23 rd, 2004, for example, ABC 揥orld News Tonight� closed with a story on the final episode of 揝ex in the City,� a show about city women looking for Mr. Right. The piece by David Wright quoted the show抯 creator, Candace Bushnell, and a number of fans, most of whom were women. The next most popular category for women was the 搈iscellaneous� category, where women appeared in 50% of the coverage.

On morning news, where celebrity and lifestyle stories make up a greater portion of content, they are again the most likely places to find female voices. Nearly two-thirds (62%) of lifestyle stories and 55% of celebrity stories offered a woman抯 perspective. No other topic category reached the 50% threshold.

On the PBS 揘ewsHour� and on cable news, not a single topic area included a female voice in at least 50% of the stories. The two top categories on the 揘ewsHour� were coverage of defense\military issues and science\technology (40% for both). Cable, on the other hand, followed the pattern of its commercial network counterparts, giving women the most voice in celebrity (32%) and lifestyle (26%) stories. Just 19% of government stories on cable and 15% of defense and military stories cited at least one female voice.

The War in Iraq and The 2004 Election

In the two big stories of 2004, the international war on terrorism and the 2004 elections, journalists� use of women as sources differed somewhat. In the roughly 6,300 stories about the war against international terrorism studied, women were even less likely to be a source than in the news coverage overall. A mere 20%, just two-in-ten of these stories, included a female source. Male sources, on the other hand, were cited in 79%. Journalists were slightly more likely to draw on women sources in stories about the 2004 elections (37%) though they were still more than twice as likely to offer a male perspective (88%).