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

Hypothesis testing in evolutionary inference

J F Brookfield1

  • 1Department of Genetics, University of Nottingham, Queens Medical Centre, UK.

Journal of Theoretical Biology
|April 21, 1997
PubMed
Summary

This study explores statistical independence in evolutionary comparative methods. It clarifies that observed correlations between traits do not necessarily imply causation, emphasizing the need for a population context in statistical analysis.

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

Correspondence.

Trends in ecology & evolution·2011
Same author

Forced and natural molecular evolution.

Trends in ecology & evolution·2011
Same author

Population bottlenecks.

Current biology : CB·2001
Same author

Selection on Alu sequences?

Current biology : CB·2001
Same author

An investigation of the cause of low variability on the fourth chromosome of Drosophila melanogaster.

Molecular biology and evolution·2001
Same author

Predicting the future.

Nature·2001

Area of Science:

  • Evolutionary Biology
  • Biostatistics
  • Phylogenetics

Background:

  • Comparative methods in evolutionary biology often analyze correlations between traits to test hypotheses.
  • Statistical independence is a key concept for hypothesis testing in these analyses.
  • Previous interpretations have sometimes conflated statistical correlation with causal relationships.

Purpose of the Study:

  • To examine the statistical basis for testing null hypotheses using correlations between binary evolutionary variables.
  • To define the conditions under which evolutionary traits and events can be considered statistically independent.
  • To critically evaluate the inference of causality from statistically significant correlations in comparative studies.

Main Methods:

  • Conceptual analysis of statistical independence in the context of evolutionary comparative methods.
  • Examination of the role of a population of observations in defining independence.
  • Logical assessment of the relationship between statistical significance and causal inference for correlated traits.

Main Results:

  • Statistical independence of observations is only meaningful within a defined population of possible observations.
  • Observed correlations between traits in comparative analyses do not inherently indicate a causal link.
  • Statistical significance of a correlation is neither a necessary nor a sufficient condition for inferring causality.

Conclusions:

  • A robust understanding of statistical independence is crucial for valid hypothesis testing in evolutionary biology.
  • Careful consideration of the sampling context is required when assessing trait independence.
  • Causal inferences from correlational data in comparative studies must be made with caution, avoiding overstatement.

Related Experiment Videos