When journalist Mark found a tip about a university researcher bending data to make a cancer treatment look more effective than it was, he knew he had a story worth chasing. What he did not know was that the investigation would take five months, require him to reconstruct the original data from scratch, and end with his editor asking whether the story was worth the cost. He published it anyway. But many similar stories never see the light of day.
That is one of the findings from a new study that tried to understand how science journalists actually do their jobs when they act as watchdogs — meaning they investigate problems in scientific research and hold scientists accountable. Led by Alice Fleerackers, a media scholar at the University of Amsterdam, the research team interviewed 21 science journalists across Canada and the United Kingdom between September 2024 and June 2025. Their work was published in the journal Science Communication.
The study found that watchdog science journalism is far more labor-intensive than most readers probably realize. Fleerackers put it plainly: "This work requires much more than a single phone call to an expert and a quick quote." Journalists described investigations that took weeks or even months, involving dense scientific papers, reconstructed datasets, and conversations with sources who were sometimes afraid to speak up.
The challenge is that this careful work often clashes with how modern newsrooms operate. Audiences want new content fast, and algorithms tend to reward simple, attention-grabbing stories over deeply reported investigations. Meanwhile, budget cuts have left many newsrooms without the legal support, fact-checkers, or specialized editors they would need to confidently publish accusations against researchers — accusations that could lead to lawsuits. The result, the study found, is that even research integrity issues that clearly deserved public attention were quietly dropped.
This matters more than it might seem. Millions of scientific articles are published every year, and only a fraction involve fraud, data manipulation, or other integrity problems. But when flawed studies get media coverage before they are corrected — or without ever being corrected — people can form wrong ideas about health risks, environmental threats, or the safety of new treatments. That erodes public trust in science as a whole.
Fleerackers argues that protecting this kind of journalism requires concrete action. She says media organizations need to give reporters more time, training, and institutional support to do this work properly. Funding bodies and universities could help by offering grants specifically for investigative science journalism and by cooperating more openly with reporters who come asking questions.
"When research integrity issues are carefully investigated and explained, it helps both researchers and the public better understand what can go wrong and how these problems can be addressed," Fleerackers said. The hope is that, by making that work more feasible, more stories like Mark's will actually get told.