CultureBERT: Measuring Corporate Culture With Transformer-Based Language Models
This provides a more accurate tool for researchers and practitioners analyzing corporate culture, though it is incremental as it applies existing transformer methods to a new domain-specific task.
The paper tackles the problem of measuring corporate culture from employee reviews by fine-tuning transformer-based language models on a unique human-labeled dataset, achieving a 17 to 30 percentage point improvement in classification accuracy compared to traditional methods.
This paper introduces transformer-based language models to the literature measuring corporate culture from text documents. We compile a unique data set of employee reviews that were labeled by human evaluators with respect to the information the reviews reveal about the firms' corporate culture. Using this data set, we fine-tune state-of-the-art transformer-based language models to perform the same classification task. In out-of-sample predictions, our language models classify 17 to 30 percentage points more of employee reviews in line with human evaluators than traditional approaches of text classification. We make our models publicly available.