Over a million developers have joined DZone.
{{announcement.body}}
{{announcement.title}}

Is Your Favorite Movie Biased? Let the Data Decide

DZone's Guide to

Is Your Favorite Movie Biased? Let the Data Decide

Cold, hard data tells a different story to what we notice at face value, teasing out a perspective that has enormous implications for the industry.

· Big Data Zone ·
Free Resource

Hortonworks Sandbox for HDP and HDF is your chance to get started on learning, developing, testing and trying out new features. Each download comes preconfigured with interactive tutorials, sample data and developments from the Apache community.

Ever notice how the best movie roles always seem to go to guys? Or that a huge proportion of the dialog assigned to female characters is just… talking about men? Think about your favorite film for a second. How many female characters are even in it? Do they say anything? Do they talk to any other women? Do they discuss anything other than their male love interests?

Hmm… now you come to think of it, you’re not sure, right?

Now, you’ve probably never really considered whether that movie you love is intrinsically sexist, but chances are, it is in super subtle ways that you don’t even notice while you’re enjoying the snappy script or finely tuned plot. This is one of those times where cold, hard data tells a very different story to what we notice at face value, teasing out a perspective that tells us a ton about a bigger issue you might never have thought too much about but has enormous implications for the industry.

You might be thinking, But hang on, isn’t this stuff subjective? What kind of data analysis could you possibly use here to tease out insights about the way a filmmaker thinks about women and their role?

Well, actually, since 1985, a fascinating “test” has been kicking around that aims to do just that. It’s called the “Bechdel Test” and is based on an idea by American cartoonist Alice Bechdel and her friend, Liz Wallace — and it’s super easy to apply to any movie you can think of.

What Is the Bechdel Test?

The Bechdel Test is straightforward. You need to take a look at the movie script and ask three questions:

  1. Are there at least two (named) female characters in this film?

  2. Do they talk to each other (not just to the male characters)?

  3. Do they talk about anything other than men?

OK, so this is a pretty basic stuff. To pass, a female character would barely need to do more than compliment another female character on her cooking, or exchange a few words about something work-related while in the office. Hardly a blueprint for feminism, right? And yet, you’d be amazed at the number of films that fail to clear this incredibly low bar.

So…Which Films Pass the Test?

As we’ve said, considering how basic a test this is, you might expect the vast majority of films to pass the test — but you’d be wrong. Data analysis of the 7,292 films in the Bechdel analysis database shows that only 57% pass, and many of them do so by the skin of their teeth!

Let’s start with some obvious ones. Movies with strong female leads are more likely to make the cut since the sheer amount of screen time they have means there’s a greater chance they’ll manage a brief chat with another woman that isn’t about the dudes.

So. Alien sees the awesome Ripley briefly (and I do mean briefly) talk strategy with the only other female crew member. Million Dollar Baby includes a heartbreaking conversation about cash between a female boxer and her morally bankrupt Mom. In Jackie Brown, the lead character has a short exchange with another female character about dropping off the money. (And another woman compliments her on her suit. We’re scraping the barrel here.)

Meanwhile, Thelma & Louise is pretty much all about the friendship between the female leads, so we get plenty of decent conversation here. Whatever Happened to Baby Jane tracks a vicious relationship between two sisters, but it passes muster nonetheless. Oh, and in Mulan, our heroine argues about marriage and defending her family’s honor with her mother, grandmother, and matchmaker.

Then there are the ones that don’t focus on women but have well-rounded female characters anyway (whoop!). Films like Sideways, that feature two female characters talking life and wine, Leon: The Professional, where a pre-pubescent Natalie Portman is warned by an older woman about the trouble that lies ahead, and Network, where a savvy news producer convinces the female leader of an underground political group to give her access to the story.

Which Ones Don’t?

Well, in the 10 years leading up to Alice Bechdel devising the famous test, the majority of films (51.3%) failed it. That’s improved since (possibly because more filmmakers shoehorn in a line or two to avoid the embarrassment of failing!) but the results are still pretty dismal. In 2016, for example, just over a third (34%) of new releases flunked it.

Unsurprisingly, big macho blockbusters like Top Gun, Predator, and the entire Mission Impossible franchise fail the test, as well as 21 of the 25 Bond films (exceptions: From Russia With Love, For Your Eyes Only, Goldeneye, and Skyfall).

Acclaimed films are even worse… of the AFI’s top 100 films of all time, only 38 films pass the Bechdel test!

More recently, films as varied as Café Society, The Social Network, Toy Story 1 and 2, The Lord of the Rings trilogy, The Avengers, The Little Mermaid, Finding Nemo, The Lion King, Aladdin, Breakfast at Tiffany’s, and Up all fail to feature a single conversation between two women that isn’t about a man.

Films That Surprisingly Do Pass the Test

That said, sometimes movies surprise you. Remember the Titans might be all about a male sports team, but some cute conversations between two young daughters about dolls and football means it passes the test. Goodfellas and American Pie 2 don’t exactly offer the healthiest portrayals of women or male-female relationships, but the former includes a short chat between two wives about Florida and the latter a superficial conversation about clothes… so officially, they tick the boxes, too.

And then there’s the whole caboodle of Disney Princess movies — Cinderella, Sleeping Beauty, Beauty and the Beast, etc. These are all ostensibly about aspiring to marry a prince, but many manage to squeeze in a few lines of dialog that isn’t about him with another female character, so they pass.

Films That Surprisingly Don't Pass the Test

On the other end of the scale are the films that feature a well-written, independent female character with plenty of depth… but simply doesn’t get screen time with any other women.

Take the awesome movie Run Lola Run: in her frantic rush to save the day, Lola barely has a proper conversation with anyone, but the few she does talk to are men. Meanwhile, Frances McDormand’s portrayal of a rural detective in Fargo might be an iconic female role, but the only women she talks to are witnesses discussing male suspects.

Meanwhile, in the original Star Wars Trilogy, Avatar, and Harry Potter and the Deathly Hallows Part II, plucky female characters might take the matter into their own hands, but they only discuss their plans with the boys.

When Pure Data Isn’t Enough

This is where things get really interesting. Data tells a story — in many cases, a story you weren’t expecting — offering up insights about a bigger narrative that we don’t expect. At the same time, it often takes a human eye to assess the results, analyzing the root causes, identifying anomalies and bringing a common sense approach that helps make sense of what we’re seeing.

In the case of the Bechdel Test, that means interpreting whether the content of what is said is a better indicator of whether a film really is sexist that who it is said to. For example, it’s pretty obvious to most people that a female character talking about shoes or babies probably won’t do much to challenge stereotypes, whereas the competent female leader of a male team very well might. In this case, the Bechdel test gives us an idea of whether a film is meeting some basic equality criteria, but the rest can be misleading.

In fact, it’s this relationship between pure data analysis and human interpretation that often shows up the flaws in preset parameters, shows us how we need to adapt and adjust queries or helps to question whether we’re using the right data sources to answer the question at hand.

After all, when The Blind Side (which earned Sandra Bullock a well-deserved Oscar) is rated below Scream 3 in terms of its feminist credentials, you know you might need to tweak your methods a little!

Hortonworks Community Connection (HCC) is an online collaboration destination for developers, DevOps, customers and partners to get answers to questions, collaborate on technical articles and share code examples from GitHub.  Join the discussion.

Topics:
big data ,data analytics ,bechdel test

Published at DZone with permission of

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}