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

Why do we still use stepwise modelling in ecology and behaviour?

Mark J Whittingham1, Philip A Stephens, Richard B Bradbury

  • 1Division of Biology, School of Biology and Psychology, University of Newcastle, Newcastle Upon Tyne, UK. m.j.whittingham@ncl.ac.uk

The Journal of Animal Ecology
|August 23, 2006
PubMed
Summary

Stepwise multiple regression, despite its known flaws, is still common in ecological studies. This method can produce unreliable results, as demonstrated by an analysis of yellowhammer distribution data.

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

Tropical insect biodiversity loss from forest-pasture conversion is substantially underestimated across spatial scales.

Current biology : CB·2026
Same author

Demographic causes and social consequences of adult sex ratio variation.

Nature communications·2026
Same author

Assessment of tick populations associated with capybaras in natural reserves and human-modified environments with or without invasive plants in the state of São Paulo, Brazil.

Experimental & applied acarology·2026
Same author

Nine changes needed to deliver a radical transformation in biodiversity measurement.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same author

From data to decisions: Toward a Biodiversity Monitoring Standards Framework.

Proceedings of the National Academy of Sciences of the United States of America·2026
Same author

Evolution of competitiveness during wheat domestication.

Current biology : CB·2026

Area of Science:

  • Ecology
  • Statistics
  • Behavioral Science

Background:

  • Stepwise multiple regression (SMR) is widely used in scientific research.
  • Despite established statistical drawbacks, SMR remains prevalent in ecological and behavioral journals.
  • Common issues include biased parameter estimation and multiple hypothesis testing.

Purpose of the Study:

  • To highlight the statistical pitfalls of stepwise multiple regression.
  • To demonstrate the unreliability of SMR using a real-world ecological dataset.
  • To advocate for alternative statistical approaches.

Main Methods:

  • Analysis of papers published in leading ecological and behavioral journals in 2004.
  • Worked example using a 4-year yellowhammer distribution dataset.

Related Experiment Videos

  • Comparison of results from stepwise regression versus information-theoretic (IT) analysis.

    • Main Results:

    • 57% of studies using multiple regression employed stepwise methods.
    • SMR yielded significant models yearly, but these models varied substantially and contradicted full dataset analysis.
    • IT analysis revealed numerous equally plausible models, underscoring the issue of relying on a single 'best' model.

    Conclusions:

    • Stepwise multiple regression is statistically flawed and can lead to erroneous ecological inferences.
    • The reliance on a single 'best' model from SMR is inappropriate given data complexity.
    • Information-theoretic approaches offer a more robust framework for ecological modeling and inference.