Misrepresentation of Overall and By-Gender Mortality Causes in Film Using Online, Crowd-Sourced Data: Quantitative Analysis

Abstract

Background: The common phrase "representation matters" asserts that media has a measurable and important impact on civic society's perception of self and others. The representation of health in media, in particular, may reflect and perpetuate a society's disease burden.

Objective: In this study, for the top 10 major causes of death in the United States, we aimed to examine how cinematic representation overall and by-gender mortality diverges from reality.

Methods: Using crowd-sourced data on over 68,000 film deaths from Cinemorgue Wiki, we employ natural language processing techniques to analyze shifts in representation of deaths in movies versus the 2021 National Vital Statistics Survey top 10 mortality causes. We parsed, stemmed, and classified each film death database entry, and then categorized film deaths by gender using a specifically trained gender text classifier.

Results: Overall, movies strongly overrepresent suicide and, to a lesser degree, accidents. In terms of gender, movies overrepresent men and underrepresent women for nearly every major mortality cause, including heart disease and cerebrovascular disease (chi-square test, P<.001); 73.6% (477/648) of film deaths from heart disease were men (vs 384,866/695,547, 55.4% in real life) and 69.4% (50/72) of film deaths from cerebrovascular disease were men (vs 70,852/162,890, 43.5% in real life). The 2 exceptions for which women were overrepresented are suicide and accidents (chi-square test, P<.001), with 39.7% (945/2382) deaths from suicide in film being women (vs 9825/48,183, 20.4% in real life) and 38.8% (485/1250) deaths from accidents in film being women (vs 75,333/225,935, 33.5% in real life).

Conclusions: We discuss the implications of under- and overrepresenting causes of death overall and by gender, as well as areas of future research.

Keywords: NLP; data science; gender; media representation; mortality; natural language processing.

Figure 1.. Overall number of total deaths in Cinemorgue versus number of recorded deaths in the United States, National Vital Statistics Survey 2021 by cause of mortality. Of total film deaths, suicide and accidents are overrepresented; all other causes of death are underrepresented in film. The identity line depicted the boundary between over- and underrepresentation; causes of death above the line are overrepresented, and causes of death beneath the line are underrepresented in film.

Figure 2.. Top 10 mortality causes by gender in film versus real life. Count and proportion of deaths by gender (blue for men, mauve for women) are shown by mortality cause for film and for real life. Women suicide deaths and accidents are overrepresented in film. Men are overrepresented in all other causes of death. We only included causes of death present in the Cinemorgue dataset.