Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Expected linking error resulting from item parameter drift among the common Items on Rasch calibrated tests.

G Edward Miller1, Paul Randall Gesn, Jourania Rotou

  • 1Division of Student Assessment, Texas Education Agency, 1701 N. Congress Avenue, Austin, TX, USA. gmiller@tea.state.tx.us

Journal of Applied Measurement
|February 11, 2005
PubMed
Summary

Estimating linking constants in state assessments depends on item parameter drift. Excluding drifted items prevents linking errors when drift is external; including them prevents errors when drift is due to examinee achievement changes.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

New Linked Employee-Employer Data Show Workforce Composition Is Associated With Health Insurance Offers Among Small Employers.

Medical care research and review : MCRR·2024
Same author

The Impact of Sharing Drug Rebates at the Point of Sale on Out-of-Pocket Payments for Enrollees in Employer-Sponsored Insurance.

Value in health : the journal of the International Society for Pharmacoeconomics and Outcomes Research·2022
Same author

Net Spending On Retail Specialty Drugs Grew Rapidly, Especially For Private Insurance And Medicare Part D.

Health affairs (Project Hope)·2020
Same author

Decomposing changes in the growth of U.S. prescription drug use and expenditures, 1999-2016.

Health services research·2019
Same author

Despite Coverage Gains, One-Third Of People In Small-Firm Low-Income Families Were Uninsured In 2014-15.

Health affairs (Project Hope)·2018
Same author

High-Deductible Health Plan Enrollment Increased From 2006 To 2016, Employer-Funded Accounts Grew In Largest Firms.

Health affairs (Project Hope)·2018

Area of Science:

  • Educational Measurement
  • Psychometrics
  • Statistics

Background:

  • State assessment programs utilize Rasch-based common item linking.
  • Item difficulty parameter drift can affect linking accuracy.
  • Drifted items remain scorable despite exclusion from linking sets.

Purpose of the Study:

  • To analyze the impact of including/excluding drifted common items in linking constant estimation.
  • To determine conditions under which linking errors occur based on the source of item parameter drift.

Main Methods:

  • Theoretical derivation of the expected effects of item inclusion/exclusion on linking constants.
  • Analysis of two scenarios: drift due to external factors vs. drift due to examinee achievement changes.

Related Experiment Videos

Main Results:

  • Excluding drifted items prevents linking errors when drift is external.
  • Including drifted items prevents linking errors when drift is due to examinee achievement changes.
  • Excluding items when drift is due to achievement changes introduces linking errors.

Conclusions:

  • The source of item parameter drift is critical for accurate linking constant estimation.
  • Excluding or including drifted items strategically mitigates linking errors in state assessments.
  • Understanding drift origins is key to maintaining score comparability across assessments.