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

A warning about median reaction time.

J Miller1

  • 1Department of Psychology, University of California, San Diego, La Jolla 92093.

Journal of Experimental Psychology. Human Perception and Performance
|August 1, 1988
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

The importance of arginine 102 for the substrate specificity of Escherichia coli malate dehydrogenase.

Biochemical and biophysical research communications·1992
Same author

High-efficiency gene transfer into cardiac myocytes.

Nucleic acids research·1992
Same author

Brain-derived neurotrophic factor augments rotational behavior and nigrostriatal dopamine turnover in vivo.

Proceedings of the National Academy of Sciences of the United States of America·1992
Same author

Differentiated practice: a three-dimensional model.

Nursing management·1992
Same author

Use of unlicensed assistive personnel in acute care settings.

The Journal of nursing administration·1992
Same author

Effect of a human immunodeficiency virus protease inhibitor on human monocyte function.

AIDS research and human retroviruses·1992

Sample medians can overestimate population medians in reaction time studies, especially with skewed data and small sample sizes. This bias can distort comparisons between experimental conditions with unequal trials.

Area of Science:

  • Cognitive Psychology
  • Psychometrics
  • Experimental Psychology

Background:

  • Reaction time (RT) distributions are often positively skewed.
  • The median is a common measure of central tendency for RT data.
  • Potential biases in RT measures can impact research findings.

Purpose of the Study:

  • To investigate the bias in sample medians when estimating population medians from skewed reaction time distributions.
  • To determine the factors influencing the magnitude of this median bias.
  • To highlight the practical implications for comparing reaction times across experimental conditions.

Main Methods:

  • Computer simulations were used to model reaction time distributions with varying degrees of skew.
  • Sample medians were computed from simulated data with different sample sizes.

Related Experiment Videos

  • The difference between sample medians and population medians was analyzed.

    • Main Results:

    • Sample medians significantly overestimate population medians with positively skewed RT distributions.
    • The degree of overestimation increases with greater skew and decreases with larger sample sizes.
    • Simulations showed potential overestimations approaching 50 ms under specific conditions.

    Conclusions:

    • The median reaction time measure is biased when used with skewed distributions, particularly with small sample sizes.
    • Using sample medians to compare conditions with unequal trial numbers can lead to artifactual or concealed differences.
    • Researchers should exercise caution when interpreting median RT differences, especially in studies with unequal sample sizes.