Quantifying Gender Stereotypes in Japan between 1900 and 1999 with Word Embeddings
🤖 AI Summary
This study quantifies the century-long evolution (1900–1999) of gender stereotypes in Japanese society. Methodologically, it trains annual Japanese word embedding models to construct a dynamic embedding sequence and introduces a “gender stereotypicity score” based on cosine similarity to systematically measure temporal changes in semantic associations between gender and roles across four domains: family, work, politics, and occupation. Contrary to the conventional “role substitution” hypothesis, results reveal a co-evolutionary pattern wherein women’s linguistic representation encompasses multiple concurrent identities: stereotypic associations strengthen significantly in work and politics, while persisting—and even intensifying—in the family domain. Moreover, occupational gender stereotypicity correlates strongly with actual gender distribution in those occupations (r > 0.8), confirming tight coupling between linguistic representation and social reality. This work contributes a scalable computational framework for analyzing historical sociosemantic change.
Application Category
📝 Abstract
We quantify the evolution of gender stereotypes in Japan from 1900 to 1999 using a series of 100 word embeddings, each trained on a corpus from a specific year. We define the gender stereotype value to measure the strength of a word's gender association by computing the difference in cosine similarity of the word to female- versus male-related attribute words. We examine trajectories of gender stereotype across three traditionally gendered domains: Home, Work, and Politics, as well as occupations. The results indicate that language-based gender stereotypes partially evolved to reflect women's increasing participation in the workplace and politics: Work and Politics domains become more strongly female-stereotyped over the years. Yet, Home also became more female-stereotyped, suggesting that women were increasingly viewed as fulfilling multiple roles such as homemakers, workers, and politicians, rather than having one role replace another. Furthermore, the strength of female stereotype for occupations positively correlate with the proportion of women in each occupation, indicating that word-embedding-based measures of gender stereotype mirrored demographic shifts to a considerable extent.
Problem
Research questions and friction points this paper is trying to address.
Quantifying gender stereotypes evolution in Japan 1900-1999
Measuring gender association strength using word embeddings
Analyzing stereotype changes across Home, Work, Politics domains
Innovation
Methods, ideas, or system contributions that make the work stand out.
Used 100 yearly word embeddings for gender analysis
Defined gender stereotype value via cosine similarity differences
Applied embedding-based measures to track demographic shifts
🔎 Similar Papers
No similar papers found.
