A Textual Analysis of US Corporate Social Responsibility Reports
Bibliographic record
Abstract
We employ computer‐based textual analysis to examine disclosure patterns for a sample of US corporate social responsibility (CSR) reports from the period 2002–2016. Starting from 466 features commonly used in computational linguistics, our results show that the linguistics or disclosure patterns in CSR reports can be used to accurately predict the actual CSR performance type of CSR reporters. Specifically, we find that the two most commonly used disclosure characteristics, number of words and number of sentences, alone can be used to predict reporting firms’ CSR performance type with 81% accuracy. The accuracy of prediction increases to 96% when the top 50 linguistics features most relevant to firms’ CSR performance are included in the prediction model. In addition, we find that the linguistic features of CSR disclosure identified by our study are incrementally value relevant to investors even after controlling for the actual CSR performance score from the professional CSR rating agencies. This finding suggests that the linguistic features of CSR disclosure can be an important venue for capital market participants in evaluating firms’ CSR performance type, especially when professional CSR performance ratings are not available.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.015 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.007 | 0.005 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".