RefDoc
Haut

Faire une nouvelle recherche
Make a new search
Lancer la recherche


Titre du document / Document title

Confusion over measures of evidence (p's) versus errors (α's) in classical statistical testing

Auteur(s) / Author(s)

HUBBARD Raymond (1) ; BAYARRI M. J. (2) ; BERK Kenneth N. (3) ; CARLTON Matthew A. (4) ;

Affiliation(s) du ou des auteurs / Author(s) Affiliation(s)

(1) College of Business and Public Administration, Drake University, Des Moines, IA 50311, ETATS-UNIS
(2) Department of Statistics and Operations Research, University of Valencia, Burjassot, Valencia 46100, ESPAGNE
(3) Kenneth N. Berk is Professor Emeritus, Department of Mathematics, Box 4520, Illinois State University, Normal, IL 61790, ETATS-UNIS
(4) Department of Statistics, California Polytechnic State University, San Luis Obispo, CA 93407, ETATS-UNIS

Résumé / Abstract

Confusion surrounding the reporting and interpretation of results of classical statistical tests is widespread among applied researchers, most of whom erroneously believe that such tests are prescribed by a single coherent theory of statistical inference. This is not the case: Classical statistical testing is an anonymous hybrid of the competing and frequently contradictory approaches formulated by R. A. Fisher on the one hand, and Jerzy Neyman and Egon Pearson on the other. In particular, there is a widespread failure to appreciate the incompatibility of Fisher's evidential p value with the Type I error rate, α, of Neyman-Pearson statistical orthodoxy. The distinction between evidence (p's) and error (α's) is not trivial. Instead, it reflects the fundamental differences between Filsher's ideas on significance testing and inductive inference, and Neyman-Pearson's views on hypothesis testing and inductive behzvior. The emphasis of the article is to expose this incompatibility, but we also hriefly note a possible reconciliation.

Revue / Journal Title

The American statistician    ISSN  0003-1305   CODEN ASTAAJ 

Source / Source

2003, vol. 57, no3, pp. 171-182 [12 page(s) (article)] (dissem.)

Langue / Language

Anglais

Editeur / Publisher

American Statistical Association, Alexandria, VA, ETATS-UNIS  (1947) (Revue)

Mots-clés anglais / English Keywords

Neyman Pearson test

;

Neyman test

;

Error type I

;

P value

;

Error rate

;

Statistical estimation

;

Statistical theory

;

Statistical test

;

Hypothesis test

;

Significance test

;

Fisher information

;

Conditional distribution

;

Error probability

;

Error estimation

;

Multivariate analysis

;

Non parametric estimation

;

Mots-clés français / French Keywords

Test Neyman Pearson

;

Test Neyman

;

Erreur type I

;

Valeur P

;

Estimation paramétrique

;

62H15

;

62G10

;

62F03

;

Taux erreur

;

Estimation statistique

;

Théorie statistique

;

Test statistique

;

Test hypothèse

;

Test signification

;

Information Fisher

;

Loi conditionnelle

;

Probabilité erreur

;

Estimation erreur

;

Analyse multivariable

;

Estimation non paramétrique

;

Mots-clés espagnols / Spanish Keywords

Indice error

;

Estimación estadística

;

Teoría estadística

;

Test estadístico

;

Test hipótesis

;

Test significación

;

Información Fisher

;

Ley condicional

;

Probabilidad error

;

Estimación error

;

Análisis multivariable

;

Estimación no paramétrica

;

Localisation / Location

INIST-CNRS, Cote INIST : 9565, 35400011256485.0040

Nº notice refdoc (ud4) : 15018855



Faire une nouvelle recherche
Make a new search
Lancer la recherche
Bas