Text-Based Depression Models Demonstrate Strong Predictive Performance
Authors systematically reviewed text-based depression estimation studies (searching 2014–2025) and meta-analyzed 15 models from 11 studies, reporting a pooled correlation r=0.605. Embedding-based features and deep-learning architectures (r≈0.74 and r≈0.73) outperformed traditional features and shallow models, and clinician-diagnosed labels yielded higher performance than self-reports; reporting quality was positively associated with performance.
Key Points
- 1Report pooled effect size r=0.605 across 15 models, indicating a large association with depression outcomes
- 2Find embedding representations and deep learning outperform traditional features and shallow models in predictive accuracy
- 3Recommend clinician-diagnosed labels and transparent reporting to improve model reliability and practical comparability
Scoring Rationale
Systematic meta-analysis with peer-reviewed evidence and practical guidance; limited novelty beyond consolidating existing methods and benchmarks.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
