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

Fitting logistic IRT models: small wonder.

M A García-Pérez1

  • 1Departamento de Metodología, Facultad de Psicología, Universidad Complutense, Campus de Somosaguas, 28223 Madrid, Spain. miguel@psi.ucm.es

The Spanish Journal of Psychology
|January 5, 2002
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

Dentate Gyrus Somatostatin Cells are Required for Contextual Discrimination During Episodic Memory Encoding.

Cerebral cortex (New York, N.Y. : 1991)·2020
Same author

Botulinum Toxin Type A for the Treatment of Primary Hyperhidrosis: A Prospective Study of 52 Patients.

Actas dermo-sifiliograficas·2017
Same author

A translational approach from an animal model identifies CD80 as a candidate gene for the study of bone phenotypes in postmenopausal women.

Osteoporosis international : a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA·2017
Same author

Genetic testing among Spanish pediatric neurologists: Knowledge, attitudes and practices.

European journal of medical genetics·2016
Same author

Association of ACACB polymorphisms with obesity and diabetes.

Molecular genetics and metabolism·2011
Same author

The impact of chocolate on cardiovascular health.

Maturitas·2011
Same journal

Compassionate Leadership: Development and Cross-Cultural Validation of Compassion at Work-Leadership Behaviors Inventory (CAW-LBI).

The Spanish journal of psychology·2026
Same journal

Climber Ability and Differences in Psychological, Physiological and Behavioral Responses to an On-sight Lead Climb.

The Spanish journal of psychology·2026
Same journal

The Youth Physical Activity Promotion Model in Spain and Chile: Comparison of Psychological and Social Variables.

The Spanish journal of psychology·2026
Same journal

Tell me Why: The Attributional Styles at Work Questionnaire and its Relationship with Affectivity, Personality, and Motivation.

The Spanish journal of psychology·2026
Same journal

The Indirect Relationship between Prosociality in the Workplace and Employee Well-Being: Testing Multiple Mediators.

The Spanish journal of psychology·2026
Same journal

ICT Use at Work as a Double-Edged Sword: A Moderated Mediation Model of Employee Well-Being.

The Spanish journal of psychology·2026
See all related articles

Item response theory (IRT) models often use logistic functions, but this study shows they can fit data from other functions. This challenges the interpretation of standard IRT model parameters.

Area of Science:

  • Psychometrics
  • Educational Measurement
  • Statistical Modeling

Background:

  • Modern item response theory (IRT) models predominantly utilize logistic functions as their item response functions (IRFs).
  • While logistic functions ensure a unit interval range and monotonic increase, their parameter spaces lack clear, empirical interpretations in testing contexts.
  • Parameter estimation success in IRT does not inherently validate the existence of the underlying constructs being measured.

Purpose of the Study:

  • To investigate the implications of exclusively using logistic functions as IRFs in item response theory (IRT) models.
  • To demonstrate how the mechanical application of logistic functions can lead to superficial model fits and parameter estimates.
  • To explore the flexibility of logistic IRT models in fitting data generated by non-logistic item response functions.

Related Experiment Videos

Main Methods:

  • A simulation study was conducted to compare the performance of logistic IRT models.
  • Data were generated using both logistic and non-logistic item response functions (IRFs).
  • The ability of logistic IRT models to fit data from these different generating functions was assessed.

Main Results:

  • Logistic IRT models demonstrated a comparable ability to fit data generated by non-logistic IRFs as they did for data generated by logistic IRFs.
  • This suggests that logistic models can provide good statistical fits even when the underlying response processes differ substantially.
  • The study highlights potential discrepancies between model fit and the accurate representation of latent traits.

Conclusions:

  • The ubiquitous use of logistic functions in IRT may lead to an overestimation of model adequacy and misinterpretation of parameters.
  • The flexibility of logistic functions in fitting diverse data patterns warrants caution in inferring the existence of constructs.
  • The article proposes considering alternative IRFs to enhance the theoretical and practical interpretability of IRT models.