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

Computerized adaptive testing: a mixture item selection approach for constrained situations.

Chi-Keung Leung1, Hua-Hua Chang, Kit-Tai Hau

  • 1Department of Mathematics, Hong Kong Institute of Education, Hong Kong. ckleung@ied.edu.hk

The British Journal of Mathematical and Statistical Psychology
|November 19, 2005
PubMed
Summary
This summary is machine-generated.

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

Standardized Estimates of Second-Order Latent Growth Models: A Comparison of Alternative Latent-Standardization Methods.

Multivariate behavioral research·2025
Same author

Appropriate modeling of endogeneity in cross-lagged models: Efficacy of auxiliary and model-implied instrumental variables.

Behavior research methods·2025
Same author

Elite career expectations of adolescents: Popularity, gender differences, and social divides.

Journal of adolescence·2024
Same author

Linear Mixed-Effects Models for Dependent Data: Power and Accuracy in Parameter Estimation.

Multivariate behavioral research·2024
Same author

Flagging insufficient effort responses in surveys: Stopping rule to prevent insufficient or excessive removal of doubtful data.

Acta psychologica·2024
Same author

Attracting adolescents to become doctors and nurses: differential importance of personal and environmental factors in 61 economies.

Human resources for health·2023

Computerized adaptive testing (CAT) can be improved by integrating multiple stratification (MS) and maximum information (MI) item selection methods. This hybrid approach, MS-MI, enhances measurement efficiency and constraint adherence, especially when controlling item exposure.

Area of Science:

  • Psychometrics
  • Educational Measurement
  • Computerized Adaptive Testing (CAT)

Background:

  • Traditional maximum information (MI) item selection in CAT leads to unbalanced item exposure and high overlap rates.
  • Multiple stratification (MS) was proposed to improve item pool utilization by sorting items based on content, difficulty, and discrimination.
  • MS may not maintain high efficiency when non-statistical constraints are imposed on testing.

Purpose of the Study:

  • To explore the benefits of a mixture item selection approach (MS-MI) integrating MS and MI for CAT with non-statistical constraints.
  • To evaluate the performance of MS-MI compared to MS and MI in terms of measurement efficiency, item pool utilization, and constraint conformity.
  • To assess the effectiveness of MS-MI in controlling item exposure.

Main Methods:

Related Experiment Videos

  • Simulation studies were conducted to compare the performance of MS, MI, and MS-MI item selection approaches.
  • Evaluation criteria included measurement efficiency, item pool utilization, conformity to non-statistical constraints, and item exposure control.
  • The MS-MI approach combined elements of both MS and MI strategies.

Main Results:

  • MS consistently outperformed MI and MS-MI in item pool utilization across all simulation conditions.
  • MS-MI and MI approaches demonstrated higher measurement efficiency and better conformity to constraints compared to MS.
  • The MS-MI approach showed superior performance over MI on all criteria when item exposure control was a factor.

Conclusions:

  • The MS-MI approach offers a balanced solution for computerized adaptive testing, particularly when dealing with non-statistical constraints and the need for controlled item exposure.
  • Integrating MS and MI strategies enhances overall testing efficiency and adherence to specific testing requirements.
  • MS-MI represents a promising advancement in item selection methodology for adaptive testing environments.