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

Inductive reasoning and judgment interference: experiments on Simpson's paradox.

Klaus Fiedler1, Eva Walther, Peter Freytag

  • 1Department of Psychology, University of Heidelberg, Germany. kf@psychologie.uni-heidelberg.de

Personality & Social Psychology Bulletin
|July 27, 2004
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

Emotional pathways linking relative deprivation experiences to conspiracy beliefs: a mini-meta-analysis.

BMC psychology·2026
Same author

Flexible Diagnosticity in Person Impression Formation: An Integrative Framework.

Personality & social psychology bulletin·2026
Same author

"A community of unknowledge": A social-psychological model of the self-reinforcing cycle of social identity-driven willful ignorance and conspiracy beliefs.

Current opinion in psychology·2025
Same author

No memory, no effect: action based evaluative conditioning effects are modulated by contingency memory.

Cognition & emotion·2025
Same author

The unfairness we feel: How positive and aggressive affect could shape relative deprivation and aggression.

BMC psychology·2025
Same author

Pitting base rate driven heuristics against conditional reasoning in multivariate contingency assessment.

Journal of experimental psychology. Learning, memory, and cognition·2025
Same journal

Metacognitive and Interpersonal Intellectual Humility Are Asymmetrically Associated with Well-Being.

Personality & social psychology bulletin·2026
Same journal

Intergroup Contact and Belonging Among Ethiopian Jews in Ethiopia.

Personality & social psychology bulletin·2026
Same journal

A Taxonomy of Data Synthesis.

Personality & social psychology bulletin·2026
Same journal

When and Why Beliefs About the Causes of a Policy Problem Predict Policy Support.

Personality & social psychology bulletin·2026
Same journal

Prospects of Downward Mobility Cause Status Anxiety and Life Dissatisfaction.

Personality & social psychology bulletin·2026
Same journal

Fluency as a Cue to Authenticity.

Personality & social psychology bulletin·2026
See all related articles

People often misinterpret statistical data, similar to Simpson's paradox, by overlooking crucial covariate information. This study shows how to improve understanding of complex relationships by highlighting relevant factors.

Area of Science:

  • Cognitive psychology
  • Social psychology
  • Statistical reasoning

Background:

  • Simpson's paradox illustrates how aggregate data can obscure subgroup trends.
  • Understanding trivariate relationships requires cognitive processes akin to analysis of covariance.
  • Prior research indicates challenges in reasoning about conditional probabilities and confounding variables.

Purpose of the Study:

  • To investigate how individuals reason about Simpson's paradox in the context of gender and success.
  • To identify cognitive barriers to correctly interpreting trivariate data.
  • To explore methods for improving inductive reasoning in complex statistical scenarios.

Main Methods:

  • Participants assessed gender, success, and covariate relationships in a Simpson's paradox-like scenario.

Related Experiment Videos

  • Experiments manipulated motivation and activated mental models to assess reasoning.
  • Subsequent experiments varied the salience of the covariate and gender using temporal cues.

    • Main Results:

    • Participants frequently generalized aggregate-level disadvantages to subgroup levels, failing to account for the covariate.
    • High motivation and activated mental models did not prevent this error in initial experiments.
    • Simpson's paradox understanding improved when covariate salience increased and gender salience decreased via temporal cues.

    Conclusions:

    • Cognitive biases can lead to misinterpretations of statistical data, even with high motivation.
    • Explicitly highlighting covariates and clarifying causal order is crucial for mastering Simpson's paradox.
    • Interventions focusing on covariate salience and causal temporal cues can enhance statistical reasoning abilities.