ipipbig-fivefacetsauthoring
Our approach: facet-level items inspired by public-domain IPIP patterns
How PsiMatch authors 1-5 Likert items per facet, routes them by id prefix to trait aggregates, and keeps the active MindSync instrument at 20 while the extended bank grows toward 500.
Citation
PsiMatch Product & Research Team (2026). PsiMatch Methodology
Read methodology5 min · Updated 2026-08-25
Educational context only, not a diagnosis or clinical assessment. Product-authored items are labeled where validated instruments do not apply.
§ 01 — Findings
What the study found
- We author original 1-5 Likert items (Strongly disagree → Strongly agree) where agreement always means more of the trait — no reverse framing in this tranche.
- Every id is namespaced (opn-/con-/ext-/agr-/neu-/eq-*/relatt-/relcom-/relcnf-/rellove-/val-) and the server routes by prefix before the legacy category router.
- The active MindSync set stays capped at 20-30; the extended bank currently contains 330 items across batches 01-11 (including scenario, forced-choice and semantic-differential formats), remains inactive for core scoring, and is surfaced through daily micro-questions and agent asks.
- All banks carry researchBasis: “Product-authored items inspired by public-domain IPIP facet patterns (Goldberg, 1999); not a validated clinical instrument.”
§ 02 — Encoding
How we encode it
- Makes new questions scorable on day one — no silent items.
- Enables MCP completion flows: get_completion_gaps → get_next_questions (full text + options) → submit_assessment_responses.
- Authoring skill at .claude/skills/psychometric-authoring/SKILL.md makes the 500-scale repeatable.