Weighted plausible-value analysis with Rubin's rules and replicate-weight variance estimation

Research Method · All research methods
Evidence · 1 claim (1 for) · 1 study (1 theoretical), q1 · 0 of 1 report an effect size · 1 claim rests on one study

Description

The article recommends that secondary analysts of PVs weight student observations to restore population representativeness, repeat each analysis separately across all PVs (usually 5 to 10), and aggregate results using Rubin's rules with replicate-weight variance estimation. Without these steps, analysts 'risk underestimating the amount of uncertainty in their findings', which can lead to overstated claims.

Accounts

• Account for both sampling and imputation error when analyzing plausible values, using weights and Rubin's rules: The article recommends that secondary analysts of PVs weight student observations to restore population representativeness, repeat each analysis separately across all PVs (usually 5 to 10), and aggregate results using Rubin's rules with replicate-weight variance estimation. Without these steps, analysts 'risk underestimating the amount of uncertainty in their findings', which can lead to overstated claims. (Dumas et al. (2026))

Claims

• Plausible values in large-scale assessment datasets are model-imputed draws from a posterior distribution, not individual test scores [+W]

Related Research Methods

-

Key Sources

• 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

Page metadata

Field	Value
type	research-method
id	weighted-plausible-value-analysis-with-rubins-rules-and-replicate-weight-variance-estimati
title	Weighted plausible-value analysis with Rubin's rules and replicate-weight variance estimation
description	The article recommends that secondary analysts of PVs weight student observations to restore population representativeness, repeat each analysis separately across all PVs (usually 5 to 10), and aggregate results using Rubin's rules with replicate-weight variance estimation.
status	review
generated	by process:settle-candidates on 2026-10-19
trust	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.