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

Basic statistics and the inconsistency of multiple comparison procedures.

David J Saville1

  • 1Statistics Group, AgResearch, Lincoln, New Zealand. dave.saville@agresearch.co.nz

Canadian Journal of Experimental Psychology = Revue Canadienne De Psychologie Experimentale
|November 5, 2003
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 ecological fitness of the tomato potato psyllid after transferring from non-crop host plants to tomato and potato.

PloS one·2022
Same author

The ability of the green peach aphid (<i>Myzus persicae</i>) to penetrate mesh crop covers used to protect potato crops against tomato potato psyllid (<i>Bactericera cockerelli</i>).

PeerJ·2020
Same author

Non-host larvae negatively impact persistence of the entomopathogen Beauveria bassiana in soil.

Journal of invertebrate pathology·2018
Same author

Host Plant Selection by the Wheat Bug, Nysius huttoni (Hemiptera: Lygaeidae) on a Range of Potential Trap Plant Species.

Journal of economic entomology·2018
Same author

Agroecological management of a soil-dwelling orthopteran pest in vineyards.

Insect science·2016
Same author

Potential ecosystem service delivery by endemic plants in New Zealand vineyards: successes and prospects.

PeerJ·2016

This study clarifies statistical measures like pooled standard deviation and least significant difference (LSD). The unrestricted LSD procedure is recommended for generating hypotheses in experimental analysis.

Area of Science:

  • Statistics
  • Experimental Design
  • Data Analysis

Background:

  • Statistical measures are crucial for interpreting experimental results.
  • Understanding interrelationships between measures like pooled variance (MSE), standard error of the mean (SEM), and standard error of the difference (SED) is essential.
  • Multiple comparison procedures (MCPs) aid in pairwise mean comparisons when no prior hypotheses exist.

Purpose of the Study:

  • To summarize and illustrate the interrelationships of common statistical measures used in scientific journals.
  • To compare various multiple comparison procedures (MCPs) for their consistency in analyzing experimental data.
  • To recommend the most suitable MCP for hypothesis generation in experimental research.

Main Methods:

  • Graphical illustration of relationships between pooled standard deviation, MSE, SEM, SED, and least significant difference (LSD).

Related Experiment Videos

  • Comparison of several well-known MCPs using a fictitious dataset.
  • Evaluation of MCPs based on the consistency of results between overall and subset analyses.

    • Main Results:

    • The least significant difference (LSD) is identified as the most useful measure among SEM, SED, and LSD for graphical representation.
    • The unrestricted least significant difference (LSD) procedure demonstrated superior consistency in analyzing both overall and subset experimental data.
    • The unrestricted LSD procedure is recommended for hypothesis generation.

    Conclusions:

    • The unrestricted LSD procedure is a valuable tool for generating hypotheses in subsequent experimentation.
    • Statistical measure selection and understanding their relationships are key to robust data interpretation.
    • MCPs, particularly the unrestricted LSD, enhance the exploratory phase of scientific research.