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

Statistical methods for analysing discrete and categorical data recorded in performance analysis.

Alan M Nevill1, Greg Atkinson, Mike D Hughes

  • 1School of Sport, Performing Arts and Leisure, University of Wolverhampton, UK. a.m.nevill@wlv.ac.uk

Journal of Sports Sciences
|October 5, 2002
PubMed
Summary

Statistical analysis of discrete performance indicators requires methods beyond normal distribution. Log-linear and logit models are more effective than chi-square tests for complex categorical data in sports performance 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

Time Series Response Analyser v2.0: A Web-Based Tool for Transparent Summary Statistics From Discrete Time-Series Data.

International journal of sport nutrition and exercise metabolism·2026
Same author

Individual oxidative stress and inflammation responses to vitamin C supplementation: Aggregated sets of n-of-1 trials.

Free radical biology & medicine·2026
Same author

Misconceptions about variance comparisons for exploring the presence of treatment response heterogeneity in randomized controlled trials.

Journal of applied physiology (Bethesda, Md. : 1985)·2026
Same author

Musculoskeletal Injuries and Illness in Professional British Tennis Players: An Injury Surveillance Study.

International journal of sports physical therapy·2025
Same author

The physical and physiological demands of intercounty Camogie match-play between competitions.

Journal of sports sciences·2025
Same author

Stature is the key: A systematic review and meta-analysis on the role of stature and body mass in physical fitness through allometric modeling.

Journal of science and medicine in sport·2025

Area of Science:

  • Sports Science
  • Statistical Modeling
  • Performance Analysis

Background:

  • Discrete events in sports performance do not follow a normal distribution.
  • Traditional statistical methods may not adequately capture the nuances of categorical data in sports.
  • Performance indicators are often discrete variables requiring specialized analytical approaches.

Purpose of the Study:

  • To identify and compare appropriate statistical methods for analyzing categorical differences in discrete performance indicators.
  • To evaluate the effectiveness of chi-square tests versus log-linear and logit models for notational analysis.
  • To demonstrate the superiority of log-linear and logit models for complex categorical data and binary response variables in sports.

Main Methods:

  • Comparison of two statistical approaches: chi-square tests (goodness-of-fit, independence) and log-linear/logit models.

Related Experiment Videos

  • Application of methods to examples from notational analysis in sports.
  • Utilizing statistical software GLIM for fitting log-linear and logit models.

    • Main Results:

    • Both chi-square tests and log-linear/logit models yield similar results for simple one-way and two-way comparisons.
    • Log-linear and logit models are more effective for complex models and higher-order comparisons.
    • Logit models are essential for analyzing binomial or binary response variables in sports performance.

    Conclusions:

    • Log-linear and logit models provide greater insight into the mechanisms of sport performance when analyzing discrete events from notational analysis.
    • For binary outcomes like shot success or goal scoring, logit models are the most appropriate statistical tool.
    • The study advocates for the adoption of advanced statistical models for robust sports performance analysis.