Breaking up is hard to do: the heartbreak of dichotomizing continuous data.
D. Streiner. Canadian journal of psychiatry. Revue canadienne de psychiatrie, 47 (3):
262-6(April 2002)5206<m:linebreak></m:linebreak>LR: 20071115; JID: 7904187; CIN: Can J Psychiatry. 2002 Nov;47(9):886. PMID: 12500760; CIN: Can J Psychiatry. 2003 Jul;48(6):429-30; author reply 430. PMID: 12894624; ppublish;<m:linebreak></m:linebreak>Anàlisi de dades; Categorització.
Abstract
Researchers often take variables that are measured on a continuum and then break them into categories (for example, above or below some cut-point), either to place subjects into groups or as an outcome measure. In this article, we show that the rationales given for this practice are weak and that categorization results in lost information, reduced power of statistical tests, and increased probability of a Type II error. Dichotomizing a continuous variable is justified only when the distribution of that variable is highly skewed or its relation with another variable is nonlinear.
Canadian journal of psychiatry. Revue canadienne de psychiatrie
number
3
pages
262-6
volume
47
city
Department of Psychiatry, University of Toronto, Kunin-Lunenfeld Applied Research Unit, Baycrest Centre for Geriatric Care, Toronto, ON. dstreiner@klaru-baycrest.on.ca
%0 Journal Article
%1 Streiner2002a
%A Streiner, David L
%D 2002
%J Canadian journal of psychiatry. Revue canadienne de psychiatrie
%K Bias(Epidemiology) ClinicalTrialsasTopic ClinicalTrialsasTopic:statistics&numericald DataInterpretation Humans OutcomeAssessment(HealthCare) OutcomeAssessment(HealthCare):statistics&num PsychiatricStatusRatingScales PsychiatricStatusRatingScales:statistics&num Psychometrics Statistical
%N 3
%P 262-6
%T Breaking up is hard to do: the heartbreak of dichotomizing continuous data.
%U http://www.ncbi.nlm.nih.gov/pubmed/11987478
%V 47
%X Researchers often take variables that are measured on a continuum and then break them into categories (for example, above or below some cut-point), either to place subjects into groups or as an outcome measure. In this article, we show that the rationales given for this practice are weak and that categorization results in lost information, reduced power of statistical tests, and increased probability of a Type II error. Dichotomizing a continuous variable is justified only when the distribution of that variable is highly skewed or its relation with another variable is nonlinear.
%@ 0706-7437
@article{Streiner2002a,
abstract = {Researchers often take variables that are measured on a continuum and then break them into categories (for example, above or below some cut-point), either to place subjects into groups or as an outcome measure. In this article, we show that the rationales given for this practice are weak and that categorization results in lost information, reduced power of statistical tests, and increased probability of a Type II error. Dichotomizing a continuous variable is justified only when the distribution of that variable is highly skewed or its relation with another variable is nonlinear.},
added-at = {2023-02-03T11:44:35.000+0100},
author = {Streiner, David L},
biburl = {https://www.bibsonomy.org/bibtex/2feb2d4b062cf22773608b32ed201a8d3/jepcastel},
city = {Department of Psychiatry, University of Toronto, Kunin-Lunenfeld Applied Research Unit, Baycrest Centre for Geriatric Care, Toronto, ON. dstreiner@klaru-baycrest.on.ca},
interhash = {07bd0cd4f628a77f2c140f24d7fcf1b1},
intrahash = {feb2d4b062cf22773608b32ed201a8d3},
isbn = {0706-7437},
issn = {0706-7437},
journal = {Canadian journal of psychiatry. Revue canadienne de psychiatrie},
keywords = {Bias(Epidemiology) ClinicalTrialsasTopic ClinicalTrialsasTopic:statistics&numericald DataInterpretation Humans OutcomeAssessment(HealthCare) OutcomeAssessment(HealthCare):statistics&num PsychiatricStatusRatingScales PsychiatricStatusRatingScales:statistics&num Psychometrics Statistical},
month = {4},
note = {5206<m:linebreak></m:linebreak>LR: 20071115; JID: 7904187; CIN: Can J Psychiatry. 2002 Nov;47(9):886. PMID: 12500760; CIN: Can J Psychiatry. 2003 Jul;48(6):429-30; author reply 430. PMID: 12894624; ppublish;<m:linebreak></m:linebreak>Anàlisi de dades; Categorització},
number = 3,
pages = {262-6},
pmid = {11987478},
timestamp = {2023-02-03T11:44:35.000+0100},
title = {Breaking up is hard to do: the heartbreak of dichotomizing continuous data.},
url = {http://www.ncbi.nlm.nih.gov/pubmed/11987478},
volume = 47,
year = 2002
}