The Bechdel test is a deceptively simple measure of female representation in fiction, most commonly applied to films. To pass, a movie must meet three criteria: it has at least two named women, those women talk to each other, and their conversation is about something other than a man. The test has become one of the most widely recognized shorthand tools for spotting gender imbalance in storytelling, yet roughly 40 percent of American films still fail it.
A Comic Strip That Became a Cultural Standard
The test originated in a 1985 comic strip called “Dykes to Watch Out For” by cartoonist Alison Bechdel. In one strip, a character explains that she only watches a movie if it satisfies those three conditions. Bechdel herself has credited her friend Liz Wallace with the idea, and the test is sometimes called the Bechdel-Wallace test. The strip was a joke, not a rigorous framework for film criticism, and Bechdel has said as much in interviews over the years. But the joke resonated because it named something audiences had felt without articulating: a huge number of mainstream movies feature women who exist only in relation to male characters, or who never speak to one another at all.
What made the test stick was its low bar. It does not ask whether a film is feminist, whether its female characters are well-written, or whether women drive the plot. It asks only whether two named women have a single conversation that is not about a man. The fact that so many films cannot clear even this threshold is the point. The test was never meant to evaluate individual movies so much as to expose a pattern across the industry.
How Many Films Pass and How That Has Changed
In the United States, about 40 percent of films fail the Bechdel test, according to research analyzing the crowdsourced database bechdeltest.com, where users rate thousands of movies against the three criteria.1PDXScholar. Testing the Bechdel Test That means a slight majority pass, but the number that fail is striking given how minimal the requirements are. A film can pass with a single brief exchange between two women about the weather and still technically clear the bar.
The picture has improved over time, though slowly. A large-scale data science study that built an automated Bechdel test classifier and applied it across decades of films found a general trend of growth in the probability of passing over time.2Humanities and Social Sciences Communications. Using data science to understand the film industry’s gender gap Films from recent decades are more likely to pass than those from the mid-twentieth century, but progress has been uneven and genre-dependent rather than a smooth upward line.
Genre Matters More Than You Might Expect
Not all genres treat female characters the same way, and the data reflects this clearly. The same study that tracked trends over time also analyzed how network-based gender features varied across genres. Genres including horror, mystery, musical, music, history, film noir, and war showed the most similar distributions of character interaction features between men and women, meaning the structural gap in how male and female characters are connected within those films was smallest.3Humanities and Social Sciences Communications. Using data science to understand the film industry’s gender gap
Some of these results feel intuitive. Horror films often center on female protagonists and feature ensemble casts where women interact with each other naturally. Musicals tend to have large casts with multiple female roles. But others are surprising: war films, a genre typically associated with male-dominated casts, showed relatively balanced interaction structures. The explanation may be that war films that do include women tend to give them substantive roles rather than decorative ones, but the research does not pin down a single cause.
Action and adventure films, by contrast, tend to show larger gender gaps in character networks. These are also among the highest-grossing genres globally, which means the films with the widest audience reach are often the ones with the weakest female representation by this measure.
The Box Office Argument
One of the most persistent industry assumptions has been that films centered on women, or films with strong female representation, earn less money. The Bechdel test has provided a way to test that assumption with data, and the results push back hard against it. A study examining the relationship between Bechdel test passage and box office revenue found that including just one conversation between two female characters was associated with roughly 23 percent higher domestic box office revenue in North America and about 27 percent higher revenue internationally.4Journal of Screenwriting. The economic case for equality in screenplays: The Bechdel test, female dialogue and box office revenue
These are not small numbers. The finding directly contradicts the belief that audiences, especially international ones, are less interested in films where women talk to each other. The researchers noted that their results indicate a significant economic benefit to increasing female on-screen portrayals, which reframes the conversation from a moral argument to a financial one. Studios that avoid female-driven stories out of commercial caution may be leaving money on the table.
It is worth being careful about causation here. Films that pass the Bechdel test may share other qualities that drive revenue, like better writing or more diverse storytelling. The association does not prove that adding a single conversation between women causes a revenue bump. But it does demolish the idea that female representation is a box office risk.
Who Writes and Directs the Story
Whether a film passes the Bechdel test is not just about casting. It correlates strongly with who is making the film. Research examining the relationship between behind-the-camera roles and on-screen representation found that female filmmakers, and female screenplay writers in particular, are instrumental for movies to have better female representation. Yet the percentage of female filmmakers in the industry has been very low.5ACM/IMS Transactions on Data Science. Measuring Female Representation and Impact in Films over Time
This finding creates a feedback loop that is easy to see but hard to break. When women are underrepresented among writers and directors, films are less likely to feature substantive female characters. When those films dominate the market, studios use their commercial performance as evidence for what audiences want, which further marginalizes female-driven stories and the women who would create them. The Bechdel test, in this context, is measuring not just what ends up on screen but a structural imbalance in who gets to tell stories at all.
What the Test Cannot Measure
The Bechdel test’s simplicity is both its greatest strength and its most serious limitation. Because it evaluates only dialogue and character names from a script, it cannot account for the full range of gender bias in a visual medium. A study on image-based analysis of gender representation in commercial films noted that the Bechdel test is incapable of considering the broad spectrum of bias that films can have as a visual medium, since its analysis is based solely on the script.6Proceedings of the ACM on Human-Computer Interaction. Quantification of Gender Representation Bias in Commercial Films based on Image Analysis
Think about what that means in practice. A film could pass the Bechdel test while still framing its female characters through a male gaze, giving them minimal screen time, or placing them exclusively in decorative or sexualized visual contexts. Conversely, a film could fail the test and still feature a complex, well-drawn female protagonist who simply happens to be the only woman in the story. “Gravity,” the 2013 film starring Sandra Bullock as an astronaut fighting to survive alone in space, is a classic example: it fails the Bechdel test because there is essentially only one female character, yet few would call it a film with poor female representation.
Other blind spots are more subtle. The test treats all passing conversations equally. A two-second exchange where one woman asks another for directions counts the same as a film built around the deep friendship between two female leads. It also cannot distinguish between a film where two women discuss their careers at length and one where they briefly discuss shoes. The binary pass/fail structure flattens these differences entirely.
Alternative Tests That Try to Go Further
Frustration with the Bechdel test’s limitations has produced several alternative measures, each trying to capture a different dimension of female representation. A feminist analysis of Chinese women-centered films used three evaluation tools side by side: the Bechdel test, the Sexy Lamp test, and the Mako Mori test, along with feminist film theory, to examine how mainstream cinema constructs female subjectivity.7Proceedings of the 3rd International Conference on Social Sciences and Cultural Studies. Beyond Empowerment Narratives: A Feminist Analysis of Three Representative Chinese Women-Centered Films
Each test targets a different failure mode:
- Sexy Lamp test: Could you replace the female character with a sexy lamp and have the plot still work? If yes, the character has no real agency. This test catches films where women are technically present but functionally inert.
- Mako Mori test: Named after the character in “Pacific Rim,” this asks whether the film has at least one female character with her own narrative arc that is not about supporting a man’s story. It addresses the Bechdel test’s blind spot around solo female leads.
- DuVernay test: Named after filmmaker Ava DuVernay, this asks whether characters of color have fully realized lives rather than existing to prop up white characters. It extends the representation lens beyond gender.
None of these tests, including the Bechdel test itself, works well in isolation. A film could pass all of them and still be mediocre in its treatment of women, or fail most of them and still offer something interesting. Researchers and critics increasingly use them as a battery rather than relying on any single metric, which gives a more textured picture of where representation falls short.
Sweden’s A-Rating Experiment
In 2013, Sweden became the first country to incorporate the Bechdel test into its film rating system. Four Swedish cinemas, backed by the Swedish Film Institute, began labeling films with an “A” rating if they passed all three criteria. The campaign attracted international attention and sparked considerable debate about whether a rating system should attempt to measure gender representation at all.8Journal of Scandinavian Cinema. After the A-rating mo(ve)ment: The Bechdel test in Swedish screen culture and beyond
The initiative was sometimes described as a movement, and it popularized the Bechdel test concept in Sweden and, through media coverage, around the world. Critics argued that the rating oversimplified representation and could mislead audiences into thinking a passing film was inherently progressive. Supporters countered that visibility was the point: the A-rating was not meant to certify quality but to make the pattern of underrepresentation impossible to ignore. Even if you disagreed with the metric, you now had to reckon with the fact that most Hollywood blockbusters could not meet three very basic criteria.
The Swedish experiment did not spread widely to other national rating systems, but it influenced how film festivals, distributors, and cultural institutions think about representation labeling. It demonstrated that a simple, legible standard can shift public conversation even when the standard itself is imperfect.
Automating the Test at Scale
One of the more interesting recent developments is the effort to automate Bechdel testing using computational tools. The crowdsourced database bechdeltest.com, where volunteers rate films, has been the primary data source for most academic studies. But human rating is slow, subjective in borderline cases, and limited to films someone has actually watched and bothered to submit.
Researchers have built automated classifiers that can assess films based on script analysis and character interaction networks. One such classifier, trained on labeled data from the crowdsourced database, achieved strong predictive accuracy and found that the most important features for predicting whether a film passes were related to the structure of character interaction triangles, meaning the patterns of who talks to whom.9Humanities and Social Sciences Communications. Using data science to understand the film industry’s gender gap This is a more sophisticated approach than simply counting lines of dialogue: it maps the social network of the film and looks at how women are positioned within it.
The Bechdel framework has also been applied to evaluate gender bias in AI-generated content. A recent study audited the screenplays produced by large language models using an automated version of the test, examining the network structure of AI-written scripts to see whether the same gender imbalances present in human-written films are reproduced by AI.10arXiv. Do Language Models Pass the Bechdel Test? Auditing Gender Biases in LLM-Generated Screenplays Since language models learn from existing text, including decades of screenplays with poor female representation, there is a real risk that AI tools will perpetuate those patterns unless specifically designed to counteract them.
The Test Beyond Film
Although the Bechdel test was conceived for movies, its logic transfers easily to other narrative forms, and researchers have applied it to literature, television, video games, and even the process of adapting books to screen. A study examining 40 film adaptations used computational methods to align book narratives with their screen versions and found that the adaptation process itself introduces gender representation issues reflective of the Bechdel test, meaning that even when a source novel features substantive female interactions, those interactions may be trimmed or eliminated in the screenplay.11ACL Anthology. Analyzing Film Adaptation through Narrative Alignment
This is a revealing finding because it isolates the screenwriting process as a point where representation shrinks. The raw material may already contain what the test asks for, but the choices made during adaptation, what to keep, what to cut, what to restructure, systematically reduce the presence of female characters talking to each other. It suggests that the gender gap the Bechdel test identifies is not just a product of original story ideas but of institutional habits in how stories are reshaped for the screen.
In television, the test faces different challenges. A single episode might fail while a full season passes easily, and serialized storytelling gives female characters more time to develop relationships that meet the criteria. The test has also been informally applied to video games, comic books, and theater, though academic research in those areas is less developed. The underlying question, whether women in a story exist as full participants in the narrative rather than as satellites orbiting male characters, translates across mediums even when the specific three-criteria format does not map perfectly.
Why a Flawed Test Still Matters
Academics, critics, and filmmakers have spent years pointing out the Bechdel test’s shortcomings, and most of those critiques are valid. It is a blunt instrument. It misses nuance. It can be gamed easily. A film could pass by having two women exchange a single line about coffee and otherwise be deeply sexist. These are real problems with the test as a measure of any individual film’s quality.
But the test was never designed to evaluate individual films. Its power is statistical. When you apply it to thousands of movies across decades and find that roughly two out of five cannot meet three trivially low criteria, that aggregate pattern tells you something the test’s per-film failures cannot obscure. The conversation it has sparked, about who speaks in stories, who they speak to, and what they are allowed to talk about, has shaped how a generation of writers, producers, and audiences think about representation. Studios now track Bechdel passage rates internally. Film databases include it as a searchable metric. Researchers use it as a starting point for more sophisticated analyses of gender in media.
The test’s cultural staying power, nearly four decades after a joke in a comic strip, suggests it captured something that more rigorous tools have struggled to replace. It is easy to understand, easy to apply, and the results are consistently uncomfortable for an industry that considers itself progressive. That discomfort is doing useful work, even when the tool itself is imperfect.