News interviews from expert sources’ perspective

One of the more consistent (and, sadly, unsurprising) findings in news research around the world is the dominance of men as sources in news stories. Numerous studies have examined the ratio of men to women as sources over the years, including two multinational studies this decade, and have found a fairly consistent proportion of 25-30% women sources across most types of media and parts of the world. This imbalance is especially pronounced among expert sources, where  studies have found about 20-25% of the experts cited are women.

What’s a bit harder to definitively tease out is why this imbalance persists. Reporters certainly gravitate to expert sources they’ve talked to before, and those go-to experts tend to be men. This has been found to be the case even when journalists say they actively try to include women as sources.

But are there factors on the source’s side that are driving this as well? Are women more reluctant to serve as sources because sexism in interviews produces negative experiences? Are they less likely to be put forward by their employers because of institutional sexism? Or perhaps they’re more likely to face gendered harassment as a result of appearing in media? These were the questions put forward by an Australian team of researchers led by Kathryn Shine in a recent study, “Does Gender Influence Media Engagement? Attitudes and Experiences of Australian Media Sources,” published in Journalism Practice.

Shine and her colleagues surveyed 220 subject experts and spokespeople, about two-thirds of them women and most of them Australian, about their experiences with media interviews, as well as their willingness and motivations to participate in those interviews, and their barriers to participation as well. Many of their respondents had quite a bit of experience working with journalists, but others had done only a few interviews.

They’ve got good news and bad news for us; let’s start with the good news. Both men and women generally had very positive experiences with interviews, and were extremely willing to be interviewed. In rating their willingness to be interviewed, they gave an average of more than 6 on a 7-point scale. Both men and women reported similar amounts of harassment as a result of their media appearances, and neither generally appeared to be deterred by it.

Now the bad news: The most common barriers to giving interviews among both men and women were concerns about being misquoted, distrust of journalists or media outlets, and lack of time. But women were significantly more likely than men to cite concerns about potential harassment and their appearance, as well as lack of specific experience, as barriers to agreeing to interviews.

For men, several of these barriers — lack of confidence, concern about being misquoted, and concern about their English-speaking competence — did not actually affect their willingness to give interviews, though they did make women less likely to give interviews. In other words, even if men said they lacked confidence, it didn’t stop them from being willing to participate in interviews. Not so for women, who reported much higher levels of nervousness and lower levels of confidence than men.

This study suggests, then, that the gender imbalance in expert sourcing isn’t just a function of the gender-based ruts that reporters’ source selection tends to run in (though, based on sources’ overall eagerness and on previous research, that’s certainly a big factor). It’s also a product of deep-seated sexist cultural norms that erode the confidence of would-be women sources and may make them less willing to be interviewed.

But Shine and her colleagues note that the overall positive experiences of the women in her study as expert sources offers hope that these inequities can be overcome. Citing Shine’s earlier research on women expert sources, the authors point out that in particular, expert sources say they feel “more comfortable and confident when a journalist demonstrates knowledge of them and their area of expertise in their initial interview approach.” So in addition to a more intentionally diverse proverbial Rolodex, an interview request from a respectful journalist who’s done their homework still goes a long way.

Research roundup

“‘Or they could just not use it?’: The dilemma of AI disclosure for audience trust in news.” By Benjamin Toff and Felix M. Simon, in The International Journal of Press/Politics. Newsrooms are increasingly turning to artificial intelligence (AI) to assist in producing the news. Just this week, for example, one of the most prominent news organizations (and thus one of the most widely mimicked in the journalism field) has reportedly gone “all in” on using AI internally, at least for some tasks: “The New York Times is greenlighting the use of AI for its product and editorial staff, saying that internal tools could eventually write social copy, SEO headlines, and some code.”

In making the case for using AI and introducing a new internal tool for doing so, the Times told staff in its editorial guidelines: “Generative AI can assist our journalists in uncovering the truth and helping more people understand the world. Machine learning already helps us report stories we couldn’t otherwise, and generative AI has the potential to bolster our journalistic capabilities even more.”

But as AI becomes more fully incorporated into news products, across many news organizations and including in the Global South, a key question remains: Should news outlets disclose when AI has helped generate an article? While transparency is considered a foundational journalism norm and is often touted as a way to build public trust in journalism (e.g., “show your work”), this study by Toff and Simon suggests that labeling news as AI-generated may actually backfire — potentially making audiences less trusting, not more. Though there are some important qualifiers to consider.

Toff and Simon conducted a survey experiment in the U.S., testing how people reacted to actual AI-generated news articles. They found that when a news story was labeled as AI-assisted, readers rated it as less trustworthy than one without a disclosure — even though they didn’t judge the AI-written content as any less accurate or fair. Here’s the thing, though: This skepticism was most pronounced among those who already had higher levels of trust in the news and a stronger understanding of how journalism works. Meanwhile, those who were already distrustful of news showed little change in their perceptions. All of this indicates that existing doubts about the media might overshadow concerns about AI.

Yet the study uncovered a paradox: while AI disclosure may have reduced trust for some, a large share of readers still wanted that disclosure, with more than 80% of all respondents saying news organizations should label AI-generated content. This contradiction underscores the central dilemma of AI transparency — people are asking for openness, but they might not necessarily like what it reveals.

The study also revealed one potential solution, though. When AI-generated articles included a list of sources — showing where the AI had pulled its information from — the negative effect on trust largely disappeared. This suggests that while audiences are wary of AI, they might not be inherently opposed to it — so long as they feel they can verify the information themselves. Pairing AI disclosures with clear sourcing might be the way forward for news organizations seeking to cultivate audience confidence in forms of journalism that rely more and more on AI tools.

(Transparency disclosure of our own: At RQ1, we are tentatively — and carefully — experimenting with using AI to help us summarize research articles and capture key data points we might have missed, and more so for this issue than any previous one. In this month’s research roundup, Seth tested having ChatGPT 4o produce some initial paragraphs capturing a study’s most consequential findings. He found that that chatbot could do this quite effectively, when checked against the articles, even if much of the text was ultimately revised or reorganized. Perhaps more interesting, though, was how ChatGPT 4o could pinpoint a few salient data points or pithy quotes worth highlighting — and these, too, were checked for accuracy, resulting in a few changes required to get them just right. We welcome your thoughts about how you are using [or not using] AI in your own work, and whether you feel it can help or hinder what we’re trying to do with RQ1.

“Integrated newsrooms: An examination of workplace satisfaction for Black journalists.” By Sharon Bramlett-Solomon, Mia Moody, and Gheni Platenburg, in Journalism & Mass Communication Quarterly. For decades, newsrooms have pledged to improve the diversity of their ranks, yet Black journalists remain underrepresented and often face particular challenges at work. This study examined the job satisfaction of Black journalists in the U.S., uncovering the factors that shape their workplace experiences (of note: the authors say their dataset, featuring 385 Black journalists, is the largest of its kind).

One of the study’s main findings is that job satisfaction for Black journalists is linked to appreciation, competitive salaries, and newsroom innovation — not just diversity initiatives. The authors acknowledge that such factors for job satisfaction are likely to apply to journalists of all races and backgrounds, but that Black journalists may experience these things differently as well.

And indeed, many respondents reported feeling isolated, facing exclusionary behaviors, and encountering structural limitations in career advancement. Overall, 50% of the journalists surveyed said they had experienced racism at work. As one respondent put it bluntly: “Newsrooms are still not diverse enough. … We are sitting at the table in greater numbers but are expected just to shut up and eat, so to speak.”

Despite these challenges, the study also points to areas of progress: “Respondents mentioned the importance of feeling appreciated, receiving positive reinforcement, having competitive job salaries, experiencing empathetic management, fostering newsroom innovation, and investing in Black journalists to help them learn new digital-age skills.” Ultimately, the authors conclude, the study offers hope that “journalists who are happy with their jobs are more likely to have higher-quality workplace productivity and retention rates.”

Fostering real inclusion, then, may be foremost about attending to the hiring, retaining, and promoting of Black journalists.

“That s**t is hard to get away from”: Working alone in U.S. rural journalism.” By Gregory Perreault, Louisa Lincoln, Mildred Perreault, and Jessica Fargen Walsh, in Journalism Practice. Continuing the theme of job satisfaction, consider what we learn about the state of working in U.S. rural journalism from this study by Perreault and colleagues. The picture they present is a sobering one: of journalists often toiling in relative isolation, dealing with immense pressure and growing hostility, and struggling to disconnect from their job — even when their safety is on the line. Through in-depth interviews with 28 rural journalists, the study finds that many are drawn to the work by a sense of community, only to discover that “working conditions are structured to maximize their labor while minimizing the support for their labor.”

The study shows how deeply and persistently rural journalists feel the weight of their job — often at the expense of their personal lives. Unlike their urban counterparts, who work in teams, many rural journalists are their town’s only reporter, meaning that if they take a break, the news stops.

“Like rural U.S. firefighters, if [rural journalists] leave for vacation, essential services are lost,” the authors write. “And like rural firefighters, they sometimes bear the enormous weight of emotionally taxing professional experiences without the infrastructure and support offered by more metropolitan regions of the US and across the globe.”

That pressure takes a toll. One journalist described getting called to cover a fatal car crash at 3 a.m., then struggling to cope because of what they just experienced (“you can take five or six showers and the smell of decay is going to stay in your nose for some reason”). Others spoke about the inability to “turn off” their work, even during time with family, leading to strained relationships and lost family connections (“My son…told me, ‘You don’t care about doing anything with me. You only want to go if you’re covering something for work'”).

While many of those challenges have been around for some time, the new wrinkle appears to be the growing hostility that rural journalists in America face — including and most perniciously from people in power. Interviewees reported being harassed at public meetings or even physically threatened for their reporting. One journalist, who had exposed corruption, feared that if they were caught alone, “they would have found me buried in the woods.” Another was told by fire department officials at a public meeting, “If you quote one thing we say, we’re going to sue you.” The risks of reporting, especially on politically charged topics, seem higher than ever.

Despite these challenges, many rural journalists remain committed to their work, some describing it as a calling rather than just a job. But there’s a paradox: while rural journalism is often framed as a community-driven endeavor, journalists receive little real support from either their news organizations or the public. They are expected to be deeply embedded in their communities — yet find themselves isolated, overworked, and at odds with local power structures.

This poses a double whammy for rural journalists vs. their urban counterparts: “While the precarious labor conditions U.S. rural journalists report are felt more acutely compared to metropolitan U.S. reporters,” the authors write, “we find the rural journalists are in a particularly weak position to mitigate these conditions.”

Ultimately, even as many of the journalists interviewed talking about sticking it out and staying on the job, convinced that they are providing a critical public service worth fighting for, the study raises unsettling questions about the future of journalism in rural communities under such conditions.