Latent-score linking method¶
Research Method · All research methods
Evidence · 5 claims (5 for) · 1 study (1 associational),q2· 0 of 1 report an effect size · 5 claims rest on one study
Description¶
A prediction-model approach that links scores across different K-2 literacy assessments using test records already available in administrative data. As the authors describe it, "Our method of linking scores between different assessments is best described as a prediction model where the same student was administered multiple distinct tests at the same time, and across years." Unlike traditional equating, it uses pre-existing test records rather than researcher-designed administrations, leverages within-student growth across assessments, and produces latent scores from multiple sources rather than symmetrical pair conversions.
Accounts¶
- Latent-score linking method for combining multiple K-2 literacy assessments from administrative data: A prediction-model approach that links scores across different K-2 literacy assessments using test records already available in administrative data. As the authors describe it, "Our method of linking scores between different assessments is best described as a prediction model where the same student was administered multiple distinct tests at the same time, and across years." Unlike traditional equating, it uses pre-existing test records rather than researcher-designed administrations, leverages within-student growth across assessments, and produces latent scores from multiple sources rather than symmetrical pair conversions. (Allensworth et al. (2026))
Claims¶
- K-2 literacy assessments are positively correlated, with MAP showing the strongest correlations with the other assessments [+W]
- K-2 assessment and latent scores predict third grade ELA grades with correlations from 0.52 to 0.63 in pre-pandemic cohorts [+W]
- Combined latent scores are highly predictive of third grade assessment scores in pre-pandemic years, with correlations from 0.67 to 0.82 [+W]
- Post-pandemic, K-2 scores predict third grade ELA outcomes with smaller but still moderate-to-large correlations (0.30-0.64 for grades; 0.47-0.81 for IAR) [+W]
- TRC-Spanish is the assessment least correlated with the others, though strongly correlated with TRC-English [+W]
Related Research Methods¶
-
Key Sources¶
- Allensworth, E., Gordon, A. & Young, C. (2026). Combining early grade assessments to study literacy skills: Addressing the variability in tests taken across schools and students. Journal of Research on Educational Effectiveness, 19(1), 1-26. https://consortium.uchicago.edu/publications/combining-early-grade-assessments-to-study-literacy-skills. doi:10.1080/19345747.2026.2636957
Page metadata
| Field | Value |
|---|---|
type |
research-method |
id |
latent-score-linking-method |
title |
Latent-score linking method |
description |
A prediction-model approach that links scores across different K-2 literacy assessments using test records already available in administrative data. |
status |
review |
moved_from |
element/latent-score-linking-method-early-grade-assessments |
generated |
by process:sweep-kinds on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — allensworth-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.