From human artefact to machine output: automating the “art” of psychological measurement
Bibliographic record
Abstract
Creating psychological assessment tools is crucial for research but traditionally expensive and time-consuming. While Large Language Models (LLMs) show promise for automating this process, existing approaches lack systematic, user-friendly methodologies grounded in psychometric principles. This study presents an enhanced Psychometric Item Generator (PIG) method using conversational LLMs with Problem-Solving Plans (PSP) and Chain-of-Thought (CoT) prompting. Three demonstrations validated the approach: Gemini 1.5 Flash generated 20 “propensity to trust AI” items with strong semantic coherence; Claude 3 Opus created 20 “AI anxiety” items that outperformed human-generated versions linguistically; and a 6-item “AI adoption in online learning” scale was developed and validated with 1,233 participants using multiverse analysis. Results demonstrate that LLMs can produce psychometrically sound items. The AI-generated anxiety scale showed superior linguistic properties compared to human alternatives, while the learning scale exhibited good internal consistency, item homogeneity, and clear two-factor structure across multiple analytical teams. The study establishes a PSP-CoT framework that improves LLM output quality, offering researchers a cost-effective, accessible scale development methodology. However, findings emphasize that human oversight, rigorous validation, and ethical considerations remain essential components of the process.
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 | 0.045 | 0.189 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.004 |
| Scholarly communication | 0.004 | 0.005 |
| Open science | 0.002 | 0.005 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.006 | 0.003 |
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".