Findings and Statistics

Nine months: Findings on women as sources, authors in Gender Check monitoring project

Jasmine R. Linabary5 Comments

Online-only news sites provide more female sources but fewer female bylines than their newspaper counterparts, at least according to our findings at the nine-month mark of our Gender Check monitoring project.

Last Friday, Oct. 21, 2011, marked nine months of our project which aims to monitor gender representations on U.S. news websites.

Through this project, we’ve been monitoring two websites — one associated with a newspaper and one that was online-only — in four different geographic regions once a week. These websites included the Seattle Times, Seattle P-I, New York Times, ProPublica, Stltoday.com, St. Louis Beacon, Miami Herald and Patch (Seminole Heights).

During each “Gender Check,” we selected the lead article from each site at the time we visited and recorded information on the gender of the author and the breakdown of the genders of the human sources referenced in the articles among other details. (For more on what Gender Checks are, read our introductory post here.)

Between the project’s start date, Jan. 18, and the conclusion of nine months, we’ve monitored a total of 272 articles. This broke down to 78 articles from the West and from the Midwest, 64 articles from the Northeast and 52 articles from the South. As we’ve stated previously, these lower counts in the Northeast and South may have some slight affect overall, but these regions’ standings among the others have remained fairly consistent over time.

Here are our findings from our first nine months:

Sourcing

Women made up 25.3 percent of human sources referenced in the articles we monitored during this time period. This matched our findings at the six-month point. In total, the articles contained 775 male sources and 263 female sources. This does not include those whose gender could not be identified, such as unnamed and unidentified sources.

Roughly 36 percent — 98 of the 272 articles monitored — contained only male sources (no female sources). In comparison, 20 articles had only female sources, or 7.4 percent. Nineteen articles contained no sources at all.

Our highs and lows still belong to the fourth and fifth months, respectively. The high came with women as 30.4 percent of sources in the fourth month followed by our low at 19.5 percent in the fifth month.

As was true at the six-month mark, the South had the highest percentage of women sources, with 33.6 percent. The lowest again belonged to the Northeast at 20.5 percent.

Here’s how that broke down geographically:

  • West: 192 males, 74 female (Women at 27.8 percent)
  • Northeast: 244 males, 63 females (Women at 20.5 percent)
  • Midwest: 248 males, 80 females (Women at 24.4 percent, the exact same as at the six-month point)
  • South: 91 male, 36 female (Women at 33.6 percent)

As we’ve found since the first quarter of our study, online-only outlets have continued to use a slightly higher percentage of female sources than those associated with a newspaper. At the nine-month mark, women were 28.3 percent of sources at online-only sites and 22.5 percent at newspaper sites. This held true in all regions except the West, where both sites had the same percentage of female sources.

  • Newspaper website: 416 males, 121 females (Women at 22.5 percent)
  • Online-only: 359 males, 142 females (Women at 28.3 percent)

The two lowest percentages of female sources are still coming from newspaper sites — the New York Times (only 13.28 percent) and Stltoday.com (at 21.3 percent). The number of articles in our sample is still low, so that limitation should be kept in mind, but this trend has held thus far.

Authorship

While source percentages appeared to hold, the percentage of female authors increased between the six-month and the nine-month mark. In total, women wrote 104 of the articles in our study, while men bylined 144. Twenty-one articles had a shared byline between a man (or men) and a woman (or two). Three articles were by staff or an unidentified author.

All of this meant women bylined 38.2 percent of the articles in our study, and 41.9 percent of articles by a person(s) of one gender or the other. This compares to 34.7 percent and 37.7 percent in our six-month findings.

Of the articles, seventeen were written by one or more man, but only one was written by more than one woman.

Our low to-date in this study came during the ninth month, when women wrote six of 28 articles, while men wrote 18. Our high came in the fourth month, when women bylined 16 of 32 stories, the only time in our study that women’s bylines exceeded those of men. Men in turn wrote 12 and the remainder were shared bylines between a man and a woman.

By geographic region. the Northeast again had the lowest percentage of female bylines at 25 percent. The South’s percentage remained fairly consistent, with women near parity at 44.2 percent.

  • West: 24 stories by an individual woman, 50 by one or more man, three with a shared byline between a man and a woman, two by contributors (Women at 30.8 percent overall)
  • Northeast: 16 by an individual woman, 37 by one or more man, 11 with a shared byline (Women at 25 percent overall)
  • Midwest: 29 by a woman, 43 by one or more man, six with a shared byline (Women at 37.2 percent overall)
  • South: 23 by a woman, 26 by one or more man, two with a shared byline, one contributor (Women at 44.2 percent overall)

In bylines, a similar trend continued that we’ve seen since our first quarter findings. Even though they have fewer female sources, newspaper websites have continued to show a higher percentage of female bylines than their online-only counterparts.

Approximately 37.5 percent of articles from newspaper websites were bylined by women. This compared to 30.1 percent at online-only outlets. This again held true in all regions, expect the Northeast. ProPublica has widened the gap from our six-month mark, with 28.1 percent female authors compared to the New York Times’ 21.8 percent. It’s also noteworthy that one newspaper website has actually turned up more female bylines than male bylines at this point — the Miami Herald — with 13 articles by females, 11 by males and two with a shared byline. Our sample size is smaller for this site, so we’ll be watching to see if this holds once it increases.

  • Newspaper website: 51 by a woman, 70 by one or more man, 13 with shared bylines between a man and woman, one by contributors (Women at 37.5 percent overall)
  • Online-only: 41 by a woman, 86 by one or more man, eight with shared bylines and two other (Women at 30.1 percent overall)

Author gender and source selection

Again, like we have in other monitoring studies, we’ve looked at how the author’s gender affected the use of female sources. At this point, female authors have a slightly higher percentage of female sources.

In the nine months of our study thus far, females were 27.2 percent of sources in articles written by women and 23.6 percent in those written by men. Articles with a shared byline between a man and women had women as 28.6 percent of sources.

Up next and review

Watch for an update on our findings on female authors and sources by article subject and on source order and gender in the coming weeks.

We’ll still be doing our weekly Gender Checks for the next three months to finish out a full year of data. Be sure to keep an eye out for our other monthly tallies as well as individual Gender Checks. Review past posts on Gender Check findings below:

We hope you remember that our study is still young and has some limitations. Further research is still needed. For other data on gender representations in online news and to compare these findings to others, check out our “Findings and Statistics” category and our “Useful Resources” page.

Related: Nine months: Science and health articles showing high female sources, low female bylines (A look at our Gender Check data by article subject)

Lowest showing of female bylines yet found in ninth month of Gender Check project

Jasmine R. LinabaryLeave a comment

During the latest month of our Gender Check monitoring project, only 25 percent of authors were women, the lowest percentage yet in this project. The only other time the percentage of female bylines dipped below 30 percent was in the third month of our study, when women bylined 26.9 percent of articles by an author of one gender or the other.

This marked the ninth month of this project, which began in January 2011. Later this week we’ll share findings from the first three quarters of our year-long endeavor.

Here’s the breakdown of our findings:

Ninth Month: Sept. 19 – Oct. 21, 2011

During the ninth month, we reviewed 28 articles, two in each Gender Check. That included five Gender Checks from the West and Midwest geographic regions and two from both the Northeast and the South.

As part of each Gender Check, our monitors look at two websites from that region — one associated with a newspaper and one that is online-only. We then select the top or lead articles on their websites at the time of the check and collect information on the author’s (or authors’) gender and the genders of the human sources referenced among other details. (For more on what Gender Checks are, read our introductory post here.)

Sourcing

This month the articles contained 26 female sources and 70 male sources, making women 27 percent of human sources whose gender could be identified. Three of this month’s articles contained no identified sources.

Here’s how sourcing broke down by geographic region:

  • West: 20 males, 9 female (Women at 31 percent)
  • Northeast: 11 males, 1 females (Women at 8.3 percent)
  • Midwest: 37 males, 11 females (Women at 22.9 percent)
  • South: 2 male, 5 female (Women at 71.4 percent)

We’ve also broken down our findings by news sites associated with a traditional newspaper and those that are not. Here are this month’s results:

  • Newspaper website: 48 males, 14 females (Women at 22.6 percent)
  • Online-only: 22 males, 12 females (Women at 35.3 percent)

Authorship

During this month, six articles were written by one or more woman and 18 by one or more man. That gave women 25 percent of bylines of one gender or another. There were also two shared bylines between a man and a woman (or two), one by staff and another that’s author was unknown.

Here’s the break down of bylines by geographic region:

  • West: 3 stories by women, 6 by men, 1 by staff
  • Northeast: None by a woman, 3 by men and 1 with a shared byline between a man and two women
  • Midwest: 2 by a woman, 7 by men, 1 with a shared byline between a man and a woman
  • South: 1 by a woman, 2 by a man, 1 unknown

Here’s how women did in bylines between newspaper sites and online-only sites this month:

  • Newspaper website: 4 by women, 9 by men and 1 with a shared byline between a man and woman
  • Online-only: 2 by a woman, 9 by men, 1 with a shared byline, 1 by staff and 1 unknown

As a reminder to our readers, these findings reflect a limited amount of data from our simple Gender Checks. We hope you recognize the limitations of this data, since we’ve only sampled a few articles from eight news sites. Further research and time is needed to verify any validity across the board.

In July, we released our findings from our first six months of monitoring. During that time, women were 25.3 percent of human sources referenced and 37.7 percent of authors of one gender or the author. Keep an eye out for findings to-date from the first three-quarters of the year later this week.

To look at past month breakdowns and other data on gender representations in online news, check out our “Findings and Statistics” category. 

Second month in a row with dismal female byline count in New Media Index stories

Jasmine R. LinabaryLeave a comment

Following a methodology change for the New Media Index reports, The Gender Report has for the second month in a row found the lowest count of female bylines yet, this month coming in at a dismal 9.5 percent.

Here at The Gender Report, we’ve been monitoring the web’s top articles based on links provided in the Project for Excellence in Journalism‘s weekly New Media Index roundups since January. PEJ’s reports the top five news stories and opinion pieces around the web in a Monday to Friday week based on commentary on blogs and social media sites. For the purposes of our study, we’ve chosen to focus on the top five in the blogosphere. When more than one link was provided on the topic, we’ve tried to monitor the first mentioned or the one that appears based on the writing to be more dominant, unless it is mentioned that articles shared the attention for that particular subject.

As we mentioned last month, August marked some changes to the New Media Index’s methodology, including the use of more sites to track the top stories as well as using a larger sample size and range of sources. (Read more about those change and the process here.) As a result, we’ve continued to notice changes in link diversity (particularly from blogs) and in topics, with more technology topics making it into the top five. As we’ve seen so far, this has meant a decrease in the number of sources and the percentage of female bylines. We’ll be continuing to monitor the differences in our results for the months following this methodology change.

Here’s what we found in the past month:

September 2011

PEJ New Media Index: September 26-30, 2011
Between Sept. 5 and Sept. 30, the New Media Index included 24 articles that could be checked. News regarding the new iPhone, usually from tech blogs, appeared every week in the top five. An apology message from Netflix CEO Reed Hastings regarding recent changes to the company’s services appeared in the top five two weeks in a row but is only counted once here. For three top stories during this time, no specific link was provided. Those topics were Facebook (No. 3, Sept 26-30), Google News (No. 3, Sept. 5-9), and Carol Bartz ousted from Yahoo (No. 4, Sept. 5-9).

In the stories we were able to monitor, here’s what we found:

  • Women were 17.6 percent of sources in this month’s articles. The articles contained 28 male sources and 6 female sources.
  • Half of the articles contained no human sources. Nearly all of those links were to blog posts with two coming from opinion columns. In addition, four articles featured only male sources.
  • Only two articles or posts were written or produced by a woman while 19 articles were by one or more man. That means women wrote only 9.5 percent of articles or posts by authors of one gender or the other. One article was by a male and female. Two were by staff. This tops last month’s eight-month low of female bylines, which was at 10 percent.

We’ll be keeping an eye on these findings in the coming months to see if we deduce other trends emerging now that the methodology for the top five articles has changed. For past months’ findings as well as other statistics on gender and the online news, visit our findings and statistics page or view our six-month recap of all of our projects and studies here.

Women see increase in bylines in eighth month of Gender Checks

Jasmine R. LinabaryLeave a comment

Women made up 24.3 percent of sources and 44.8 percent of authors in the eighth month of our Gender Check monitoring project. That reflected a rise in women authors over last month, when women had 35 percent of bylines.

Here’s the breakdown of our findings for the eighth month:

Eighth Month: Aug. 22 – Sept. 15, 2011

We reviewed 32 articles, two in each Gender Check, during our eighth month. That included four Gender Checks from each geographic region.

In each Gender Check, we look at two websites from that region — one associated with a newspaper and one that is online-only. During our monitoring, we pull the top or lead articles on their websites at the time of the check and gather details on the gender of the author, the breakdown of the genders of the human sources referenced in the articles and other details. (For more on what Gender Checks are, read our introductory post here.)

Sourcing

This month the articles contained 81 male sources and 26 female sources. This meant women made up 24.3 percent of the human sources referenced. Five articles this month contained no named sources.

Here’s how that broke out by geographic region:

  • West: 19 males, 8 female (Women at 29.6 percent)
  • Northeast: 18 males, 5 females (Women at 21.7 percent)
  • Midwest: 34 males, 6 females (Women at 15 percent)
  • South: 10 male, 7 female (Women at 41.2 percent)

Here’s the breakdown by news sites associated with a traditional newspaper and those that are not.

  • Newspaper website: 50 males, 14 females (Women at 21.9 percent)
  • Online-only:  31 males, 12 females (Women at 27.9 percent)

Authorship

This month 13 articles were written by an individual woman and 16 by one or more man. That put women at 44.8 percent of bylines of one gender or another for the month. Additionally, there were three shared bylines between a man and a woman (or two). This was one of the higher byline percentages for women in the study and an increase over the 35 percent women reached last month.

Here’s how bylines broke down by geographic region:

  • West: 3 stories by an individual woman, 5 by an individual man
  • Northeast: 2 by a woman, 5 by men and 1 with a shared byline between a man and a woman
  • Midwest: 4 by a woman, 3 by a man, 1 with a shared byline between a man and a woman
  • South: 4 by a woman, 3 by a man, 1 with a shared byline between a man and two women

Here’s how women did in bylines between newspaper sites and online-only sites this month:

  • Newspaper website: 7 by a woman, 7 by men and 2 with a shared byline between a man and woman (or two)
  • Online-only: 6 by a woman, 9 by a man, 1 with a shared byline

As a note to our readers, these findings reflect a limited amount of data from our simple Gender Checks. We hope you recognize the limitations of this data, since we’ve only sampled a few articles from eight news sites. Further research and time is needed to verify any validity across the board.

In July, we released our findings from our first six months of monitoring. During that time, women were 25.3 percent of human sources referenced and 37.7 percent of authors of one gender or the author.

To look at past month breakdowns and other data on gender representations in online news, check out our “Findings and Statistics” category. 

Lowest percentage of female authors yet in August’s New Media Index count

Jasmine R. LinabaryLeave a comment

August’s look at articles in the New Media Index produced an eight-month low in female bylines and one of the lowest percentages of female sources. Time will tell whether these numbers are related to a change in the methodology for determining the top five most discussed and linked to articles on the web.

Here at The Gender Report, we’ve been monitoring the web’s top articles based on links provided in the Project for Excellence in Journalism‘s weekly New Media Index roundups since January. These posts chronicle the top five news stories and opinion pieces around the web in a Monday to Friday week based on commentary on blogs and social media sites. We’ve chosen to focus on the top five in the blogosphere.

Aug. 1 marked some changes to the New Media Index report. PEJ updated its methodology to use more sites to track the top stories and began using a larger sample size and range of sources. (Read more about those change and the process here.)

For the purpose of our study, this has meant changes in the sources of links. During the first seven months of the year, the majority of links provided mostly came from the LA Times, Washington Post and in less frequency the BBC, and were mostly news articles with some opinion pieces (in other words, almost exclusively traditional media sources). This month we’ve noticed more diversity in the link sources, including from blogs and company’s direct postings.

When more than one link was provided on the topic, we’ve monitored the first mentioned or the one that appears based on the writing to be more dominant, unless it is mentioned that two articles shared the attention for that particular subject. This happened in higher volumes this month, resulting in a larger sample of articles overall.

In July, we released our findings from the first six months of monitoring. In that time, women were 20.2 percent of sources and had 31.3 percent of bylines (of articles by a person or several of one gender or the other, not including shared bylines between a woman and a man).

Here’s what we found specifically in the past month:

August 2011

PEJ New Media Index: August 1-5, 2011

Between Aug. 1 and Sept. 2, the New Media Index included 33 articles that could be checked. Reoccurring topics included the 2012 presidential campaign and the new iPhone — both appeared in the top five in four out of the five weeks. More than one link was included for nine topics. One link related to the iPhone in week four was no longer functioning and no link was provided for the No. 3 story of Google buying Motorola in week three.

Here’s what we found:

  • Women were 14.1 percent of sources in this month’s articles. The articles contained 67 male sources and only 11 female sources. This is the second lowest female source total we’ve seen. The low thus far came in May with women at 13.3 percent.
  • Most shockingly 18 out of the 33 linked to articles contained no human sources at all (though one or two did reference a company or an unnamed and unidentified official). Nearly all of those links were to posts or reviews as opposed to traditional news articles. In addition, eight articles featured only male sources.
  • Only three articles or posts were written or produced by a woman while 27 articles were by men. That means women wrote only 10 percent of articles or posts by authors of one gender or the other. Two were uncredited and one was by staff. That percentage of female authors fell below our previous low of 11.1 percent in May.

We’ll be keeping an eye on these findings in the coming months to see if we deduce other trends emerging now that the methodology for the top five articles has changed. For past months’ findings as well as other statistics on gender and the online news, visit our findings and statistics page or view our six-month recap of all of our projects and studies here.