Adding collateral information has little effect on ability and item parameter estimates but more effect on speed parameter estimates¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r2 · q2
Currency · newest study 2011 · checked against its sources 2026-10-20 (automated, 2 of 2 entries) · not reviewed by a person · overdue for a recheck since 2014: tool and model accuracy changes quickly

Subclaims¶

q2 i? Differences among models had little effect on theta (ability) parameter estimates for both tests. → Wang 2011 q2 i? Model differences had more effect on zeta (speed) parameter estimates than on theta estimates, and small effects on item b and lambda parameters. → Wang 2011 (2)

Evidence¶

Wang 2011¶

Wang, S., & Jiao, H. (2011). Incorporating Person Covariates and Response Times as Collateral Information to Improve Person and Item Parameter Estimations. Paper presented at the annual meeting of the National Council on Measurement in Education, New Orleans, LA. https://www.nwea.org/research/publication/incorporating-person-covariates-and-response-times-as-collateral-information-to-improve-person-and-item-parameter-estimations/

q2 · i? · associational · r2

Descriptive comparison of person parameter means, standard deviations and standard errors across models (Table 3, Figures 1-2); NS34 theta was negatively skewed, LI52 theta near normal.

"The difference among models have a little effect on theta parameter estimations for both NS34 and LI52 tests."

Wang 2011 (2)¶

Wang, S., & Jiao, H. (2011). Incorporating Person Covariates and Response Times as Collateral Information to Improve Person and Item Parameter Estimations. Paper presented at the annual meeting of the National Council on Measurement in Education, New Orleans, LA. https://www.nwea.org/research/publication/incorporating-person-covariates-and-response-times-as-collateral-information-to-improve-person-and-item-parameter-estimations/

q2 · i? · associational · r2

Same descriptive comparison extended to speed parameters (Figures 3-4) and item parameters (Table 4): "For both b- and lambda-parameters, the model effects are small."

"The difference among models have more effect on zeta parameter estimations for both NS34 and LI52 tests than that on theta parameter estimations."

Discussion¶

Related Claims¶

• Joint IRT models including collateral information (raw score, gender, ethnicity) fit better than item-only base models for both tests — related
• Scores from a different subject test can serve as collateral information: mathematics scores can improve reading item parameter estimation and vice versa — related
• Rescoring with new item parameters changed average ability estimates negligibly, with new and original scores correlated above 0.99 — related
• Computer-based testing's routine capture of item-level response times has driven increased interest in applications of response time to measuring ability and achievement — related
• Ability and speed are negatively correlated, and the reading test's ability estimates are more affected by speed than the math test's — related