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

Estimating person locations from partial credit data containing missing responses.

R J De Ayala1

  • 1Department of Educational Psychology, Teachers College Hall 114, University of Nebraska-Lincoln, Lincoln, NE 68588-0345, USA. rdeayala2@unl.edu

Journal of Applied Measurement
|June 30, 2006
PubMed
Summary

When handling missing responses in assessments, using the midpoint score or a likelihood-based approach is best for estimating respondent ability. Ignoring or treating omitted responses as incorrect can lead to inaccurate proficiency estimates.

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

A Comparative Evaluation of Kernel Equating and Test Characteristic Curve Equating.

Applied psychological measurement·2018
Same author

An introduction to mixture item response theory models.

Journal of school psychology·2017
Same author

The Assessment of Dimensionality for Use in Item Response Theory.

Multivariate behavioral research·2016
Same author

A Validation and Reliability Study of the Physical Activity and Healthy Food Efficacy Scale for Children (PAHFE).

Health education & behavior : the official publication of the Society for Public Health Education·2007
Same author

The effect of missing data on estimating a respondent's location using ratings data.

Journal of applied measurement·2003

Area of Science:

  • Educational measurement
  • Psychometrics
  • Data analysis in assessments

Background:

  • Assessment situations often yield partial credit data.
  • Examinees may omit responses, creating missing data.
  • Accurate estimation of respondent location is crucial.

Purpose of the Study:

  • To investigate strategies for handling missing responses in partial credit data.
  • To compare the efficacy of different missing data handling methods.
  • To determine the best approach for estimating examinee proficiency.

Main Methods:

  • A simulation study was conducted.
  • Partial credit and generalized partial credit models were used.
  • Expected a posteriori (EAP) ability estimation was applied.

Related Experiment Videos

Main Results:

  • The Midpoint and Likelihood-based procedures demonstrated superior performance.
  • Ignoring omitted responses led to less accurate estimates.
  • Treating omitted responses as incorrect also resulted in suboptimal proficiency estimation.

Conclusions:

  • The Midpoint and Likelihood methods are recommended for handling missing responses in EAP estimation.
  • Practitioners should avoid ignoring or treating omitted responses as incorrect.
  • Effective handling of missing data is vital for accurate examinee proficiency assessment.