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

Informative presentation of summary data

H T Davies1

  • 1Department of Management, University of St Andrews, Fife.

Hospital Medicine (London, England : 1998)
|November 3, 1998
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

A randomized comparative crossover study to assess the affect on circuit life of varying pre-dilution volume associated with CVVH and CVVHDF.

The International journal of artificial organs·2008
Same author

Improving care or professional advantage? What makes clinicians do audit and how well do they fare?

Health bulletin·2003
Same author

Introducing "ERIC", a living research ethics database.

Journal of medical ethics·2003
Same author

A rapidly expanding lesion in the neck: unusual presentation of Castleman's disease.

International journal of oral and maxillofacial surgery·2001
Same author

How to present the findings of research.

Hospital medicine (London, England : 1998)·2001
Same author

Exploring the pathology of quality failings: measuring quality is not the problem--changing it is.

Journal of evaluation in clinical practice·2001
Same journal

Management of non-thrombolysable acute coronary syndrome (ACS).

Hospital medicine (London, England : 1998)·2005
Same journal

Case report: treatment of faecaloma of the colon.

Hospital medicine (London, England : 1998)·2005
Same journal

Should anaesthetists scrub? The case for and against.

Hospital medicine (London, England : 1998)·2005
Same journal

Angiosarcoma: a difficult diagnosis.

Hospital medicine (London, England : 1998)·2005
Same journal

Pitfalls in recurrent varicose vein surgery.

Hospital medicine (London, England : 1998)·2005
Same journal

Aorto-caval fistula.

Hospital medicine (London, England : 1998)·2005
See all related articles

Researchers often misuse mean and standard deviation to summarize study data. This article highlights why medians and percentiles are better for describing study subjects, especially for ordinal data.

Area of Science:

  • Biostatistics
  • Medical Research Methodology
  • Data Presentation

Background:

  • Research studies frequently present summary statistics for participant data.
  • Commonly, mean and standard deviation are used to describe variables.
  • This practice can lead to inaccurate or misleading data interpretation.

Purpose of the Study:

  • To identify common pitfalls in the statistical description of study subjects.
  • To advocate for more appropriate statistical methods in research reporting.
  • To guide researchers in accurately presenting summary data.

Main Methods:

  • Review of common statistical practices in research publications.
  • Analysis of the suitability of different statistical measures (mean, standard deviation, median, percentiles) for various data types.

Related Experiment Videos

  • Discussion of the implications of inappropriate data summarization.

    • Main Results:

    • Mean and standard deviation are often inappropriately applied, particularly to ordinal data.
    • Ordinal data require non-parametric summary statistics for accurate representation.
    • Even interval data may be better represented by medians and percentiles.

    Conclusions:

    • The use of mean and standard deviation for summarizing all data types is a significant pitfall in research reporting.
    • Medians and percentiles offer more informative and accurate summaries, especially for ordinal and sometimes interval data.
    • Adopting appropriate statistical descriptors is crucial for the valid interpretation of study findings.