Race- and gender-based under-representation of creative contributors: art, fashion, film, and music

Artistic creation benefits individuals and society, encompassing economic, social, and health advantages. A crucial aspect often overlooked is the ethical imperative and practical benefits of demographic diversity within creative fields. This study addresses the scarcity of empirical evidence on the underrepresentation of minority groups in creative arts. The research focuses on influential levels of four domains: contemporary art (major U.S. museums), high fashion (runway shows), film (box office hits), and popular music (top charts). The study’s contributions include a comparative analysis across fields, examination of intersecting identities (gender and race/ethnicity), comparison with U.S. population demographics to highlight underrepresentation, and a transparent, repeatable methodology. The study's limitations, including its reliance on a convenience sample and socially inferred demographic data, are acknowledged.

Existing research on diversity in the creative arts is reviewed across the four selected domains. In contemporary art, studies reveal significant underrepresentation of women and artists of color in museum collections and auctions. Fashion shows a similar pattern, with underrepresentation of models and designers from minority groups. Studies on film highlight the lack of diversity among actors, directors, and crew, along with the perpetuation of racial biases. The music industry faces concerns about gender and racial/ethnic inequality at corporate and artist levels, with underrepresentation of women across roles and pay discrepancies.

The study analyzed over 4700 creative contributors. Data for contemporary artists was drawn from a previous study (Topaz et al., 2019). Data for fashion, film, and music were collected using Amazon Mechanical Turk (MTurk), employing a crowdsourcing approach. Experienced U.S.-based MTurk workers inferred gender and race/ethnicity of contributors based on publicly available information. To ensure data quality, multiple workers independently assessed each individual, providing confidence ratings. A six-point confidence threshold determined the final inference for each characteristic. The American Community Survey (ACS) data served as a benchmark for comparison. The methodology acknowledges limitations in relying on socially inferred demographics instead of self-reported data and the inherent challenges in defining and classifying gender and race/ethnicity. Data cleaning and standardization techniques were employed to improve data consistency and accuracy. Data validation was also performed by a researcher not involved in the initial inference process. The study uses socially inferred demographic characteristics, acknowledging limitations of this approach compared to self-reported data. Data was restricted to gender and race/ethnicity due to the difficulties in inferring other identities from publicly available sources.

The study found significant underrepresentation of women across all four creative domains: contemporary art (28%), fashion (45%), film (27%), and music (17%), compared to their 51% share in the U.S. population. Marginalized racial/ethnic groups (excluding Black musicians) were also substantially underrepresented in art (22%), fashion (22%), and film (19%), comprising roughly half their 39% share in the U.S. population. Black musicians showed a higher (48%) representation but this does not indicate equity and inclusion. White men were overrepresented in all domains (factor of 1.4-2 compared to U.S. population), while other gender-racial/ethnic groups experienced further minoritization. Proportional area charts reveal that each domain is dominated by white individuals and/or men. A bubble chart compares the intersection of gender and race/ethnicity across domains with U.S. population data, highlighting the overrepresentation of white men and the underrepresentation of nearly all other marginalized groups.

The findings demonstrate a pervasive lack of diversity across influential positions in four major creative sectors. The underrepresentation of women and marginalized racial/ethnic groups is significant and consistent across domains, suggesting systemic issues. The overrepresentation of white men indicates a power structure that continues to marginalize other groups. While higher representation of Black musicians in popular music is noted, it does not negate the broader issue of systemic inequality. The reliance on socially inferred data limits the ability to fully capture the complexities of identity. The study's findings provide quantitative evidence supporting anecdotal observations of underrepresentation, emphasizing the need for targeted interventions.

This study provides a first comprehensive, comparative analysis of gender and race/ethnicity in four major creative domains. Significant underrepresentation of women and marginalized racial/ethnic groups was found, along with overrepresentation of white men. While the study's reliance on inferred demographic data presents limitations, it provides crucial quantitative evidence of systemic biases. Future research should prioritize collecting self-identified demographic data to improve accuracy and deepen understanding. Targeted interventions and public pressure are needed to address these issues and foster more equitable representation.

The study's reliance on socially inferred demographics, rather than self-identified data, is a major limitation. The methodology might not fully capture the complexities of identity, particularly regarding gender and multiracial identities. The selection of creative domains and proxies for influential contributors represents a convenience sample, limiting the generalizability of the findings. The American Community Survey data used for comparison is also limited in its representation of gender and ethnic categories.