Bias and RMSE of ability estimates are larger for on-grade item pools than for ±1-grade and all-grade pools across content areas and simulee types¶
Claim · All claims
Evidence · 1 study · causalr2·q2
Currency · newest study 2019 · not yet checked against its sources · not reviewed by a person · overdue for a recheck since 2022: 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.
q2 i? In mathematics simulations, biases for on-grade simulees were consistently larger with the on-grade pool (ranging from -0.4 to 1) and average RMSE dropped from 2.53 (on-grade) to 2.39 (±1-grade) and 2.36 (all-grade). → Li 2019
Evidence¶
Li 2019¶
Li, S., & Meyer, J. P. (2019). Simulation study for evaluating MAP® Growth™ grade-level item pools. Portland, OR: NWEA. https://www.nwea.org/research/publication/simulation-study-for-evaluating-map-growth-item-pools-with-grade-level-constraints/
q2 · i? · causal · r2
Bias and RMSE analyses (Appendix B, Table B.1) comparing final MLE ability estimates against true RIT in the mathematics simulations. For off-grade simulees, average RMSE decreased to 2.5 for the ±1-grade pool and 2.4 for the all-grade pool versus 2.88 for the on-grade pool.
"biases for on-grade simulees are consistently larger when using an on-grade item pool (ranging from -0.4 to 1). The RMSEs for the on-grade pool are also relatively large with an average value of 2.53."
Discussion¶
Related Claims¶
- On-grade item pools cause content balancing failures for off-grade-ability students, while all-grade pools improve balancing and language usage tests remained balanced throughout — related
- Testing students with off-grade ability using on-grade item pools substantially diminishes measurement precision compared with all-grade pools — related
- On-grade mathematics item pools yield larger SEMs and lower efficiency than all-grade pools for on-grade students, especially at the low end of the scale — related
- Restricted item pools show higher item use rates than all-grade pools, and use rates drop sharply for off-grade simulees — related
- Reading SEMs exceed those in mathematics under all pool types, and only the all-grade reading pool meets the operational SEM stopping rule on average — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
on-grade-pools-larger-bias-rmse |
title |
Bias and RMSE of ability estimates are larger for on-grade item pools than for ±1-grade and all-grade pools across content areas and simulee types |
description |
Bias and RMSE of ability estimates are larger for on-grade item pools than for ±1-grade and all-grade pools across content areas and simulee types |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2019 |
recheck_by |
2022 |
generated |
by process:wiki-ingest on 2026-10-08 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — li-2019 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.