Study: Gender Bias Leads to Lower-Rated Female Films
Average Scores on IMDB Don’t Tell the Whole Story
Though "Barbie" earned nearly $1.5 billion at the box office, its popularity has fizzled by awards season—and "Oppenheimer" leads Academy Awards nominations instead. A new UMD study on gender bias in movie reviews may explain why.
"Barbie" image courtesy of Warner Brothers; "Oppenheimer" courtesy of Universal; collage by Valerie Morgan
“Barbie” might have won the dance-off against “Oppenheimer” at the box office, but a new Robert H. Smith School of Business study may explain why the hot-pink bubble burst well before this Oscar weekend.
Audiences give female-led films lower ratings, on average, but also disagree more on the quality of those films. In most cases, those movies make less money as well, according to the paper in Strategic Management Journal.
“‘Barbie’ is an anomaly,” said Associate Professor of Management and Organization David Waguespack. “It’s a female-led, major studio film that did very well (financially), even with a nontrivial number of people who said they really disliked it.”
He and co-author Bryan Stroube Ph.D. ’15, an assistant professor at London Business School, analyzed nearly 400 million consumer ratings on IMDB for 4,000 films released between 1992 and 2018. They found that men and women rated male-led movies similarly—but when it came to female-led films, male audiences were more likely to give them the worst scores, pulling down the mean.
“If the average is driven down by people giving it a ‘1’ for a movie that people otherwise like, that’s pretty misleading,” Waguespack said.
That’s what happened to “Barbie,” starring Margot Robbie, which had a big spike at the lowest rating, in contrast to the Cillian Murphy-led “Oppenheimer,” which had a more normal spread.
The differences could be ideologically driven. Some male reviewers’ “ratings may not reflect their belief about the quality of the product at all,” he said. “They just don’t like the idea of it.”
Since audience reviews may also be influenced by genre or production team, Waguespack and Stroube also used ChatGPT to generate 20 plausible movie plots with randomly assigned male and female leads and surveyed 800 people on much they thought they’d like the fake films and how eager they were to see them. The experiment confirmed that respondents were less likely to want to view female-led films.
But there’s a bright spot to this gender-typing, too. Women give more top ratings to actress-led films, and independent studios, which don’t have to appeal to the masses, earn higher box office revenue from those movies.
“If there’s a significant number of people who love a movie, it doesn’t matter if some people hate it,” Waguespack said.
Topics:
Schools & Departments: