Using Word Embeddings to Examine Gender Bias in Dutch Newspapers, 1950-1990
This addresses the problem of understanding historical media biases for researchers and historians, though it is incremental in applying existing methods to new historical data.
The paper examined gender bias in six Dutch national newspapers from 1950 to 1990 using word embeddings, finding clear differences and changes over time, such as bias shifting toward men overall despite societal trends, with increasing divergence between religious/social-democratic and liberal newspapers.
Contemporary debates on filter bubbles and polarization in public and social media raise the question to what extent news media of the past exhibited biases. This paper specifically examines bias related to gender in six Dutch national newspapers between 1950 and 1990. We measure bias related to gender by comparing local changes in word embedding models trained on newspapers with divergent ideological backgrounds. We demonstrate clear differences in gender bias and changes within and between newspapers over time. In relation to themes such as sexuality and leisure, we see the bias moving toward women, whereas, generally, the bias shifts in the direction of men, despite growing female employment number and feminist movements. Even though Dutch society became less stratified ideologically (depillarization), we found an increasing divergence in gender bias between religious and social-democratic on the one hand and liberal newspapers on the other. Methodologically, this paper illustrates how word embeddings can be used to examine historical language change. Future work will investigate how fine-tuning deep contextualized embedding models, such as ELMO, might be used for similar tasks with greater contextual information.