AI/ML Questions: AI-assisted item development and measurement
A documented AI/ML question-bank project connects AI-assisted drafting/review with intended human approval and item-level measurement/revision guidance.
L2Research workflow (workflow level, not a quality rating) Source: Repository-documented Outcome: Not empirically evaluated here
Research applicability
Concrete AI-assisted educational item development with source-described empirical item-analysis/revision method, not just generic AI topic coverage.
Getting started
Untested editorial guidance: read original source documentation and limitations, choose a suitable public/synthetic example, and plan independent source/output checks before any separately authorised trial. No trial performed.
Maintenance
Primary repository documents observed2026-10-19; future maintenance, current provider terms and live compatibility not verified.
License
README distinguishes CC BY-SA4.0 questions, MIT code and SIL fonts; licences/source rights not audited.
Workflow & human oversight
Intended L2 for AI-assisted item drafting, deterministic checks, second-model review and intended human approval, followed by documented item-statistics/revision guidance. Current alpha drafts have not all received human review. No learner-data audit, representative sampling or learning/psychometric effectiveness is established.
Workflow level is not an evidence score; higher is not better evidence.
Approach
README describes validators and a second model before human approval; statistics documentation describes difficulty/discrimination/distractor diagnostics, version filtering and reviewer decisions. Current alpha drafts are explicitly not all human-reviewed; no answer data or statistical results were inspected.
Limitations
Full README and statistics documentation read; no question contents, individual responses, database, app interaction or source code inspected/executed. Anonymous/no-cookie/no-IP data handling, counts and calculations are source claims, not audited privacy/results. Statistics guide warns self-selected users, per-browser first attempts, study-mode/order/attrition effects, version filtering and possible fabricated responses. Intended human approval must not be reported as already completed. No population generalisation, teaching effectiveness or psychometric validation claimed. AI editorial review only; no source code, models or installation executed, no production/status/publication changes.
Reproducibility
Commit-pinned primary documentation preserved; no source code/model/experiment executed or results reproduced.
Adoption requirements
Public documentation can be read without installing. A separate tool trial needs current dependency/data-rights/provider review and human question/source checking; no learner data or model calls used here.
Comments
No comments yet.
Add a comment
Comments are held for moderation before appearing. Critical or corrective comments are welcome — especially evidence that an entry overstates what has been verified.