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

Modeling data from water-level tasks: a test theoretical analysis.

Anton K Formann1

  • 1Department of Psychology, University of Vienna, Wien, Austria.

Perceptual and Motor Skills
|August 22, 2003
PubMed
Summary

This study challenges the common method of analyzing water-level task data. Latent class and Rasch models offer better descriptions of spatial ability, suggesting caution with binomial mixture models.

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

Developmental Steps in Metaphorical Language Abilities: The Influence of Age, Gender, Cognitive Flexibility, Information Processing Speed, and Analogical Reasoning.

Language and speech·2017
Same author

The covariate-adjusted frequency plot.

Statistical methods in medical research·2013
Same author

Potential predictors in neonaticide: the impact of the circumstances of pregnancy.

Archives of women's mental health·2012
Same author

Parameter recovery and model selection in mixed Rasch models.

The British journal of mathematical and statistical psychology·2011
Same author

Sensitivity to initial values in full non-parametric maximum-likelihood estimation of the two-parameter logistic model.

The British journal of mathematical and statistical psychology·2011
Same author

Pervasiveness of the IQ rise: a cross-temporal meta-analysis.

PloS one·2011

Area of Science:

  • Cognitive Psychology
  • Developmental Psychology
  • Psychometrics

Background:

  • Water-level tasks, developed by Piaget, assess spatial abilities related to liquid behavior.
  • Current analysis often dichotomizes responses and uses mixture of binomials, a method questioned here.
  • Understanding spatial reasoning development is crucial for educational and psychological assessments.

Purpose of the Study:

  • To evaluate the appropriateness of statistical models for analyzing water-level task data.
  • To compare mixture of binomials with latent class and Rasch models for spatial ability assessment.
  • To investigate the influence of age and sex on water-level task performance.

Main Methods:

  • 12 water-level tasks with round bottles at varying angles were administered to 431 participants (children and adults, male and female).
  • Statistical analyses included mixture of binomials, latent class models, Rasch model, and linear logistic test model.
  • Task difficulties and subject scores were analyzed, considering age and sex effects.

Main Results:

  • Age significantly impacted task difficulty and scores, while sex did not.
  • Mixture of binomials proved inadequate due to ignoring heterogeneity in task difficulties.
  • Latent class models successfully described data by accounting for subject and task heterogeneity.
  • The Rasch model validated a unidimensional scale for subjects and tasks, with raw scores as sufficient statistics.
  • Task difficulty parameters in the Rasch model correlated with the bottle's angle of inclination.

Conclusions:

  • Latent class and Rasch models provide more comprehensive descriptions of water-level task data than mixture of binomials.
  • The Rasch model supports a unidimensional trait interpretation of spatial ability in these tasks.
  • While not definitively resolving the trait vs. class debate, results caution against oversimplified binomial mixture analyses.
  • The angle of inclination is a key factor influencing task difficulty.

Related Experiment Videos