Content area
Full text
Contents
- Abstract
- Natural Language Processing as a Tool to Study Stereotype Change
- Expanding Across the Wider Landscape of Social Groups
- Expanding the Number of Language Corpora for Analysis
- Expanding the Metrics of Stereotype Change
- The Current Project
- Method
- Transparency and Openness
- Analysis Procedure
- Overview
- Step 1: Select the Text Corpora
- Google Books English-All Embeddings
- Corpus of Historical American English Embeddings
- New York Times Embeddings
- Common Crawl Embeddings
- Step 2: Choose Groups and Represent Groups in Labels
- Step 3: Compute Trait Associations to Groups
- Step 4: Compute Latent Averages of Valence, Warmth, and Competence
- Step 5: Compute Change in Manifest Content and Latent Structure
- Manifest Content
- Latent Structure
- Step 6: Exploring Predictors of Which Groups Change or Remain Stable
- Predicting Change From Group Types
- Predicting Change From Linguistic Features of Group Representations
- Results
- Change in Manifest Content
- Change in Latent Structure: Warmth
- Change in Latent Structure: Competence
- Change in Latent Structure: Valence
- Examining Variability in Group Stereotype Change
- Correlates of Change in Manifest Content
- Correlates of Change in Latent Warmth
- Correlates of Change in Latent Competence
- Correlates of Change in Latent Valence
- General Discussion
- Change in Manifest Content Is Greater Than in Latent Valence, Warmth, and Competence
- Change Varies in Predictable Ways Across 72 Group Stereotypes
- Correlates of Stereotype Change Across Groups: The Role of Linguistic Features
- Correlates of Stereotype Change Across Groups: The Role of Corpus Inconsistency
- Correlates of Stereotype Change Across Groups: The Role of Group Type
- Methodological Implications for Studying Group Stereotypes Through Text
- Limitations and Conclusion
Figures and Tables
Abstract
The social world is carved into a complex variety of groups each associated with unique stereotypes that persist and shift over time. Innovations in natural language processing (word embeddings) enabled this comprehensive study on variability and correlates of change/stability in both manifest and latent stereotypes for 72 diverse groups tracked across 115 years of four English-language text corpora. Results showed, first, that group stereotypes changed by a moderate-to-large degree in manifest content (i.e., top traits associated with groups) but remained relatively more stable in latent structure (i.e., average cosine similarity of top...