Abstract
The way our society is shown in films significantly shapes our perception of social normsand role models. However, multiple studies show that there is still a significant difference
in the representation of men and women in film. Although the representation of women
and men has improved in many ways over the last 50 years, we still have not reached
the point of equality.
A milestone on the topic of gender equality in films was the Bechdel test, which has
often been used to show inequality in films. However, this test has been widely criticized,
and alternative tests have been developed that promise a more accurate measurement.
This thesis explores whether there is a test that reliably measures gender equality in
films that can be performed by any layperson in a reasonable time (about one hour).
To explore this, a qualitative literature analysis was conducted to examine existing gender equality tests. Subsequently, using grounded theory, a framework for a new test
based on scientific literature was developed.
Through the qualitative literature analysis, it was found that the different tests focused
on various aspects of gender equality. None of the tests successfully addressed the topics
of gender ratio and stereotypes simultaneously. Additionally, many tests are formulated
vaguely, creating loopholes through which films that are not gender-equal could still be
considered as such. Moreover, there are hardly any scientific studies that have analyzed
these tests and their accuracy. Furthermore, almost all tests focus exclusively on women
and do not offer a corresponding alternative for men.
Inspired by scientific papers and their approaches, and utilizing Grounded Theory, a
framework for a new test was developed. Initial tests with the framework indicate that
it is capable of measuring and visualizing the gender ratio and different stereotypes
of both genders. However, even though the framework of a new test can assess gender
equality, more accurate methods of measuring gender equality are likely to emerge soon.
The first approaches to recognizing stereotypes from text using AI are already being
explored.
It can be concluded that there is currently no test that effectively measures gender
equality for non-experts. Nevertheless, it is possible to create such a test.
| Date of Award | 2024 |
|---|---|
| Original language | English (American) |
| Supervisor | Roland Keil (Supervisor) |