Behind every analysis, there is a story. 

My story behind Women & Movies is short: “Women are underrepresented in the movies.” – “That’s not true; there are so many movies with female leads!” – “Ok, name ten.” – “Ms Marvel.” And cut.

Why does it even matter?

Equal representation – of all genders, races, religions, sexualities, classes, disabilities – across the industries, including the entertainment, is essential for a healthy society. Not only it gives an example, or a character to relate to the underrepresented group of a community but also normalises their equal position in the eyes of other people, usually the majority. Role models in cultural references can lead to either strengthening the stereotypes or breaking them.

According to the latest research by the Center for the Study of Women in Television & Film cited in BBC’s article from this week, a number of female leads grew from 31% in the prior year to 40% in 2019. Forty-three per cent featured male protagonists, and 17 per cent had a combination of both sexes. That’s some good news! Well, at least for white women, who are the significant majority. Black women in speaking roles make 20%, Asian 7%, and Latinas only 5%.

Also, even though the difference between the number of movies with a female protagonist is only 3% lower (43% of movies had male lead only), their marital status was known more often (46% vs 34% for men), they were less likely to have an identifiable occupation (61% vs 73% for men) and haven’t been seen working in their work setting that often either (46% vs 59% for men).

As it might be easy for some to empathise with a group of people they are not part of, it is almost impossible to understand their life experience completely. Especially in the movie industry that often uses shortcuts and generalisation, it is vital to acknowledge intersectionality. Also, a given group’s perspective should be portrayed and worked on by a member of that group to provide an as accurate image to the audience as possible. Works of fiction out of this world are not an exception as even that world, coming out of the author’s world of imagination, is based on their experience in this world. A world where half of the population are women.

From the data point of view

As a basis, I used a data set of all movies from IMDb and selected only the top 100 of those with the highest number of reviews and top ratings. Those can be seen as the best movies, according to the (re)viewers. Additional data is coming from IMDb, Wikipedia,

Each movie has one or two directors and one or more writers as in the original IMDb data set, and three lead actors/actresses as featured on the movie’s overview page on IMDb.

As seen here, data can be misleading as Daisy Ridley, the main star of the last Star Wars trilogy is missing. IMDb lists the stars in the credits order, and Daisy Ridley is mentioned 4th there. The data used in my analysis on how many women are in the top 3 leads is dependent on the credits, therefore represents the view of the movies’ authors.

Gender was assigned to each person as of the time of making their movie (i.e. Lilly and Lana Wachowski were known as Brothers Wachowski when making The Matrix, therefore are assigned male gender in this analysis). As this data set does not include other genders (please, call me out if I’m mistaken here), the whole analysis focuses only on male/female.

All movies are also flagged as either passed or failed the Bechdel-Wallace test. A film has to have at least two (named) female characters that talk to each other about something else than a man to successfully pass the test.

The Tableau visualisation part

The analysis consists of four parts.
First part is a radar chart that focuses on movies that passed or failed the Bechdel-Wallace test and their performance against eight KPIs. Budget, income, budget/income ratio, # votes, rating, is directed by a female, is written by a female, has at least one female lead. All data is aggregated and normalised.
This chart shows that movies that pass the test do have a higher income, therefore a broader audience. They also perform well in budget/income ratio. They’re not better rated, even though the difference is 0.2 points, as the average vote for failed movies is 8.33.

Radar chart step-by-step tutorial.

The second part is a bar chart showing what genres are popular amongst female writers and well-represented by female protagonists.

The next part is a timeline of all 100 movies divided by passed/failed the Bechdel-Wallace test. The density of the circles shows when got movies with at least one female character more popular, that was in the late 1990s and around year 2010. Also, more than half of the movies in the top 100 were made after year 2000.
Within this part, additional details for each movie are included in the tooltip.

The last part focuses on the Celluloid Ceiling, film industry’s Glass Ceiling equivalent, across the whole industry in the past 40 years and within this data set.

The analysis is available on my Tableau Public profile and can be either viewed there, or downloaded as a packaged workbook. More on how to create a radar chart can be found in this post.

Leave a Comment

Your email address will not be published. Required fields are marked *