Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.02370 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917208999854080 |
|---|---|
| author | Camuffo, Arnaldo Gambardella, Alfonso Kazemi, Saeid Malachowski, Jakub Pandey, Abhinav |
| author_facet | Camuffo, Arnaldo Gambardella, Alfonso Kazemi, Saeid Malachowski, Jakub Pandey, Abhinav |
| contents | Large language models (LLMs) offer strategy researchers powerful tools for annotating text at scale, but treating LLM-generated labels as deterministic overlooks substantial instability. Grounded in content analysis and generalizability theory, we diagnose five variance sources: construct specification, interface effects, model preferences, output extraction, and system-level aggregation. Empirical demonstrations show that minor design choices-prompt phrasing, model selection-can shift outcomes by 12-85 percentage points. Such variance threatens not only reproducibility but econometric identification: annotation errors correlated with covariates bias parameter estimates regardless of average accuracy. We develop a variance-aware protocol specifying sampling budgets, aggregation rules, and reporting standards, and delineate scope conditions where LLM annotation should not be used. These contributions transform LLM-based annotation from ad hoc practice into auditable measurement infrastructure. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_02370 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Variance-Aware LLM Annotation for Strategy Research: Sources, Diagnostics, and a Protocol for Reliable Measurement Camuffo, Arnaldo Gambardella, Alfonso Kazemi, Saeid Malachowski, Jakub Pandey, Abhinav Computers and Society Computation and Language 68T07 C.4; I.2.6; I.2.7 Large language models (LLMs) offer strategy researchers powerful tools for annotating text at scale, but treating LLM-generated labels as deterministic overlooks substantial instability. Grounded in content analysis and generalizability theory, we diagnose five variance sources: construct specification, interface effects, model preferences, output extraction, and system-level aggregation. Empirical demonstrations show that minor design choices-prompt phrasing, model selection-can shift outcomes by 12-85 percentage points. Such variance threatens not only reproducibility but econometric identification: annotation errors correlated with covariates bias parameter estimates regardless of average accuracy. We develop a variance-aware protocol specifying sampling budgets, aggregation rules, and reporting standards, and delineate scope conditions where LLM annotation should not be used. These contributions transform LLM-based annotation from ad hoc practice into auditable measurement infrastructure. |
| title | Variance-Aware LLM Annotation for Strategy Research: Sources, Diagnostics, and a Protocol for Reliable Measurement |
| topic | Computers and Society Computation and Language 68T07 C.4; I.2.6; I.2.7 |
| url | https://arxiv.org/abs/2601.02370 |