Plausible values in large-scale assessment datasets are model-imputed draws from a posterior distribution, not individual test scores¶
Claim · All claims
Evidence · 1 study · theoreticalr3·q1
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q1 i? PVs are random draws from each student's posterior ability distribution based on item responses and background characteristics, making them a special case of multiple imputation rather than point-estimate scores. → Dumas 2026
Evidence¶
Dumas 2026¶
Dumas, D., Goecke, B., Kagan, S., & Weiss, S. (2026). Zooming Out On Education: Making Valid Psychological Inferences From Large-Scale Assessment Data. Educational Psychology Review. https://doi.org/10.1007/s10648-025-10110-7
q1 · i? · theoretical · r3
Conceptual analysis of the psychometric score-generation process in large-scale assessments. The article states these 'PVs are a special case of multiple imputation, where the latent proficiency is treated as a missing value and is imputed repeatedly using a conditioning model'.
"The PVs are random draws from the posterior distribution of each students' ability based on their item responses and numerous of their other observed background characteristics (Wu, 2005)."
Discussion¶
Related Claims¶
- Person-centered clustering of students on plausible values falls outside the valid use of PVs, while clustering sampling-frame units is valid — a narrower finding that bears on this claim
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
plausible-values-are-multiple-imputations-not-scores |
title |
Plausible values in large-scale assessment datasets are model-imputed draws from a posterior distribution, not individual test scores |
description |
Plausible values in large-scale assessment datasets are model-imputed draws from a posterior distribution, not individual test scores |
status |
review |
evidence_strength |
weak |
newest_evidence |
2026 |
recheck_by |
2029 |
generated |
by process:wiki-ingest on 2026-10-19 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — dumas-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.