Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Reporting Statistical Information in Medical Journal Articles

Reporting Statistical Information in Medical Journal Articles EDITORIAL Reporting Statistical Information in Medical Journal Articles TATISTICS IS not merely about distributions Much regression output serves little purpose in medi- or probabilities, although these are part of cal research publication; this usually includes the inter- the discipline. In the broadest sense, statis- cept coefficient, R , log likelihood, standard errors, and tics is the use of numbers to quantify rela- P values. Estimates of variance explained (such as R , cor- S tionships in data and thereby answer ques- relation coefficients, and standardized regression coef- tions. Statistical methods allow the researcher to reduce ficients (sometimes called effect size) are not useful mea- a spreadsheet of data to counts, means, proportions, rates, sures of causal associations or agreement and should not 9-12 risk ratios, rate differences, and other quantities that con- be presented as the main results of an analysis. These vey information. We believe that the presentation of nu- measures depend not only on the size of any biological merical information will be enhanced if authors keep in effect of an exposure, but also on the distribution of the mind that their goal is to clarify and explain. We offer exposure in the population. Because this distribution http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JAMA Pediatrics American Medical Association

Reporting Statistical Information in Medical Journal Articles

JAMA Pediatrics , Volume 157 (4) – Apr 1, 2003

Loading next page...
 
/lp/american-medical-association/reporting-statistical-information-in-medical-journal-articles-QBQtomp4YQ

References (77)

Publisher
American Medical Association
Copyright
Copyright 2003 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
ISSN
2168-6203
eISSN
2168-6211
DOI
10.1001/archpedi.157.4.321
pmid
12695224
Publisher site
See Article on Publisher Site

Abstract

EDITORIAL Reporting Statistical Information in Medical Journal Articles TATISTICS IS not merely about distributions Much regression output serves little purpose in medi- or probabilities, although these are part of cal research publication; this usually includes the inter- the discipline. In the broadest sense, statis- cept coefficient, R , log likelihood, standard errors, and tics is the use of numbers to quantify rela- P values. Estimates of variance explained (such as R , cor- S tionships in data and thereby answer ques- relation coefficients, and standardized regression coef- tions. Statistical methods allow the researcher to reduce ficients (sometimes called effect size) are not useful mea- a spreadsheet of data to counts, means, proportions, rates, sures of causal associations or agreement and should not 9-12 risk ratios, rate differences, and other quantities that con- be presented as the main results of an analysis. These vey information. We believe that the presentation of nu- measures depend not only on the size of any biological merical information will be enhanced if authors keep in effect of an exposure, but also on the distribution of the mind that their goal is to clarify and explain. We offer exposure in the population. Because this distribution

Journal

JAMA PediatricsAmerican Medical Association

Published: Apr 1, 2003

There are no references for this article.