Research Question

The Bechdel test is a test for media which can be used as a measure for the representation of women in a certain work. A piece of media passes the bechdel test if the following conditions are true:

• There are 2+ female characters that are named
• Two women have a conversation that is not about men

With this in mind, we want to see if the gross revenue of a movie affected by whether or not it passes the Bechdel test.

Our data <3

Our data contained movies made between 1970 and 2013. We chose to analyze domestic gross income, test pass/fail, and the reasons why the movies failed the test. We chose to omit entries which contained N/A in the gross income column.

## Loading required package: ggplot2

## Loading required package: dplyr

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

## Loading required package: mosaic

## Registered S3 method overwritten by 'mosaic':
##   method                           from   
##   fortify.SpatialPolygonsDataFrame ggplot2

## 
## The 'mosaic' package masks several functions from core packages in order to add 
## additional features.  The original behavior of these functions should not be affected by this.

## 
## Attaching package: 'mosaic'

## The following object is masked from 'package:Matrix':
## 
##     mean

## The following objects are masked from 'package:dplyr':
## 
##     count, do, tally

## The following object is masked from 'package:ggplot2':
## 
##     stat

## The following objects are masked from 'package:stats':
## 
##     binom.test, cor, cor.test, cov, fivenum, IQR, median, prop.test,
##     quantile, sd, t.test, var

## The following objects are masked from 'package:base':
## 
##     max, mean, min, prod, range, sample, sum

## Loading required package: kableExtra

## 
## Attaching package: 'kableExtra'

## The following object is masked from 'package:dplyr':
## 
##     group_rows

year	imdb	title	test	clean_test	binary	budget	domgross	intgross	code	budget_2013.	domgross_2013.	intgross_2013.
2013	tt1711425	21 & Over	notalk	notalk	FAIL	1.3e+07	25682380	42195766	2013FAIL	13000000	25682380	42195766
2012	tt1343727	Dredd 3D	ok-disagree	ok	PASS	4.5e+07	13414714	40868994	2012PASS	45658735	13611086	41467257
2013	tt2024544	12 Years a Slave	notalk-disagree	notalk	FAIL	2.0e+07	53107035	158607035	2013FAIL	20000000	53107035	158607035
2013	tt1272878	2 Guns	notalk	notalk	FAIL	6.1e+07	75612460	132493015	2013FAIL	61000000	75612460	132493015

Impact of Bechdel Test Result on Money Made

***

It appears that the median gross income from movies that failed the Bechdel test was slightly higher than the median gross income of movies that passed the Bechdel test. The spread of the data does appear to be similar in each group.

The Underlying data

Before conducting a t-test, we checked to see if our data was normally distributed. It was not.

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

*** However, because the two categories seemed to be similarly skewed to the right, we proceeded with the t-test.

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

T-Test

A one-way t-test was used to determine if movies that failed the Bechdel test make significantly more money than movies that failed the Bechdel test. We found that the true difference in means between the two groups was greater than zero, so the movies that failed the Bechdel test did make more money.

## 
##  Welch Two Sample t-test
## 
## data:  PassedMovies$domgross and FailedMovies$domgross
## t = -3.4618, df = 1751.1, p-value = 0.0002747
## alternative hypothesis: true difference in means is less than 0
## 95 percent confidence interval:
##      -Inf -6872092
## sample estimates:
## mean of x mean of y 
##  61885653  74985189

In which way do the movies fail the Bechdel test?

After demonstrating that movies which fail the Bechdel test make more money on average, we were curious about which of the conditions of the Bechdel test the movies failed to satisfy. It appears that the most common way to fail was “notalk”, which means that there was not a conversation between two women. Additionally, a significant proportion of the failed movies failed because it was unclear (dubious) if the conditions to pass the Bechdel test were met.

Conclusion

It seems that movies which pass the Bechdel test on average make less money than movies that fail the Bechdel test. However, a more rigorous statistical analysis which takes into account the skew of the data might yield different results. The Bechdel test is also not the most detailed or rigorous test of representation or women in media, so it might be helpful to look at more variables in the future, such as the race/ethnicty of the women in the movie.

Data Sources

-Information about Bechdel Test: https://projects.fivethirtyeight.com/next-bechdel/ -Data source: https://github.com/fivethirtyeight/data/blob/master/bechdel/movies.csvtitle: “Project1” output: ioslides_presentation —