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 methodology: II. Reliability and validity assessment in study design, Part B

D J Karras1

  • 1Division of Emergency Medicine, Temple University School of Medicine, Philadelphia, PA 19140, USA. dkarras@astro.ocis.temple.edu

Academic Emergency Medicine : Official Journal of the Society for Academic Emergency Medicine
|February 1, 1997
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

Circadian differences among 4,096 emergency department patients with acute asthma.

Critical care medicine·2001
Same author

Serum markers in the emergency department diagnosis of acute myocardial infarction.

Emergency medicine clinics of North America·2001
Same author

Clinically meaningful changes in quantitative measures of asthma severity.

Academic emergency medicine : official journal of the Society for Academic Emergency Medicine·2000
Same author

Endogenous mediators in emergency department patients with presumed sepsis: are levels associated with progression to severe sepsis and death?

Annals of emergency medicine·1999
Same author

Incidence of foodborne illnesses: preliminary data from the foodborne diseases active surveillance network (FoodNet).

Annals of emergency medicine·1999
Same author

Tick removal.

Annals of emergency medicine·1998

Assessing test validity involves comparing a test to similar measures. Quantitative analysis is possible with a reference standard, but alternative methods like interest difference techniques are useful when biases are present.

Area of Science:

  • Psychometrics
  • Measurement Science

Background:

  • Validity is crucial for evaluating the accuracy of tests and measures.
  • Traditional correlation methods like Pearson and Spearman may not fully capture validity due to potential biases.
  • Alternative approaches are needed when reference standards are unavailable or when dealing with categorical data.

Purpose of the Study:

  • To explore various methods for assessing test validity, particularly when quantitative analysis is challenging.
  • To highlight the limitations of standard correlation techniques in validity assessment.
  • To introduce alternative statistical approaches for more robust validity evaluation.

Main Methods:

  • Review of established validity concepts including criterion-based, content, and construct validity.

Related Experiment Videos

  • Discussion of correlation coefficients (Pearson, Spearman) and their limitations.
  • Exploration of interest difference techniques and the kappa statistic for categorical data analysis.
  • Consideration of questionnaire design for reliability and validity assessment using Likert scales.

    • Main Results:

    • Criterion-based validity requires a reference standard for quantitative assessment.
    • Content and construct validity may not be quantitatively assessable without a reference standard.
    • Pearson and Spearman correlations can be misleading due to unaddressed measurement biases.
    • Interest difference techniques and the kappa statistic offer more meaningful insights in specific contexts.
    • Reliability and validity of questionnaires depend on homogeneous questions and appropriate data analysis scales.

    Conclusions:

    • Validity assessment necessitates a clear definition of the test's scope and comparison with validated tools.
    • Quantitative validity assessment is enhanced by using ordinal data scales like Likert scales.
    • When reference standards are absent, alternative methods like interest difference analysis and kappa statistics are valuable for robust validity evaluation.