Ferguson also believes reporters aren't aware of other factors he's researched that could affect a study's results, like exposure to domestic violence, personality and genetics. "The psychological community in general has done a very poor job of informing people about the limitations of our research," he says, "and I think that psychological studies in general come out sounding a lot more sure of themselves than they really ought to be."
It's common practice for psychologists to identify the problems, limitations and possible errors of their own research in a discussion section at the end of a published study. "Any good scientist will tell you exactly where the flaws are in his own study," says Gentile. "We just want to get the best scientific information out to the people who can use it." While this section is an important part of the scientific process, it is rarely addressed in newspaper articles.
But if newspaper reporters aren't putting research studies in the right context or they're leaving out important information, it might not be their fault. Smaller newsrooms and bigger workloads mean one of the biggest problems may be reporters aren't even writing the stories credited to them.
"I was very surprised how many news sources pulled directly from the university's press release," says Andrew Przybylski, a psychology graduate student at the University of Rochester who led a recent group of studies on whether violent content increased players' enjoyment of a game. "I expected the articles to vary more widely based on different readings of the actual article we published."
Susan Hagen, part of the University of Rochester's communications department, says she has no problem with journalists poaching from her press releases, which she says are always checked by the researchers involved before they're sent out. Communications staff at Texas A&M International and Iowa State said they take similar precautions.
A story for Canada's financially troubled Canwest News Service about Przybylski's studies appeared in numerous newspapers across the country with a byline given to David Wylie. Most of the article is cut and pasted from the University of Rochester's press release, including direct quotes plus sentences and phrases used to describe the studies. The article never names the press release as a source, implying that Wylie spoke to Przybylski and his co-authors directly or quoted from the actual study. Attempts to contact Canwest for comment received no response.
Gentile also worries whether newspapers have become too reliant on press releases for their content. "There's a serious problem with the lack of funding for good investigative journalism, and so many papers are now relegated to just running the press release," he says. "They should make a few calls to make sure that it's credible and that the study really does show the thing that we're claiming it does. But that takes time, and that's of course the problem."
He also points out that game journalists are often guilty of the same behavior, uncritically printing positive stories about videogames based on similar press releases.
Przybylski says many of the media reports he read that were based on the press release skipped or oversimplified some important information from his research. "We were very careful to say that violence is frequently woven into the creative narratives of games," he says. "We were testing if adding more violence added to enjoyment, and if violence was a consistent motivator."