Item difficulty estimates from target-plus-adjacent-grade samples generally differ from original estimates about as much as on-grade estimates do, with mean differences within 0.11 logit¶
Claim · All claims
Evidence · 1 study · associationalr2·q2
Currency · newest study 2021 · checked against its sources 2026-10-20 (automated, 1 of 1 entry) · not reviewed by a person · overdue for a recheck since 2024: 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 Study 2 (Grades 4–7), the report says differences between three-grade and original estimates are "in general" no different from on-grade versus original differences; mean three-grade-minus-original differences ranged from −0.11 logit (Grade 5 math) to 0.01, and the 0.05 logit the report calls the largest average difference is the largest gap between the two columns (Grade 7 math). → Wei He 2021
Evidence¶
Wei He 2021¶
Wei He, NWEA Psychometric Solutions. (2021). An Investigation of Item Parameter Invariance Using Focused Calibration Samples for MAP Growth. NWEA. https://www.nwea.org/research/publication/an-investigation-of-item-parameter-invariance-using-focused-calibration-samples-for-map-growth/
q2 · i? · associational · r2
Study 2 recalibrated Grades 4–7 items with target-plus-two-adjacent-grade samples (Table 3.6, Figure 3.6). The report states "The magnitude of the largest average difference is 0.05 logit in Grade 7 math"; in Table 3.6 that figure matches the largest gap between the three-grade and on-grade mean-difference columns (Grade 7 math, −0.03 vs −0.08), while the largest three-grade-versus-original mean difference is −0.11 logit (Grade 5 math, against −0.09 on-grade), and in several rows the three-grade difference is slightly the larger. Most items' differences were within 0.3 logit, with a few math and reading items exceeding 1 logit.
"The magnitude of the largest average difference is 0.05 logit in Grade 7 math. As expected, using responses from three grades tended to increase the calibration sample size by at least two times for almost all grades."
Discussion¶
Related Claims¶
- Content Proximity pilot tests contained significantly more on-grade items and made better use of adjacent-grade items than traditional MAP Growth tests — related
- Average on-grade and all-grade item difficulty estimates differ by less than 0.1 logit for MAP Growth items — related
- For items aligned to multiple target grades, difficulty estimates from different on-grade calibration samples differ negligibly, at most 0.1 logit on average — related
- At least 90% of reading and science items and about 80% of math items show difficulty differences within 0.3 logits between on-grade and all-grade calibration — related
- Calibration sample size differences did not affect the size of on-grade versus all-grade item difficulty estimate differences — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
map-growth-three-grade-calibration-comparable-to-original |
title |
Item difficulty estimates from target-plus-adjacent-grade samples generally differ from original estimates about as much as on-grade estimates do, with mean differences within 0.11 logit |
description |
Item difficulty estimates from target-plus-adjacent-grade samples generally differ from original estimates about as much as on-grade estimates do, with mean differences within 0.11 logit |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2021 |
source_checked |
2026-10-20 |
recheck_by |
2024 |
generated |
by process:wiki-ingest on 2026-10-08 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — wei-he-2021 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.