Skip to main content
Key Specialties
Cardiology Diabetology Endocrinology Gastroenterology Geriatrics and Gerontology Gynecology and Obstetrics Hematology Infectious Disease Internal Medicine Nephrology Neurology Oncology Pediatrics Respiratory Medicine Rheumatology Surgery
Congress Coverage
2024 ACC Congress 2023 ESMO Congress 2023 EASD Congress 2023 ERS Congress 2023 ESC Congress
Clinical Collections
Advances in Alzheimer’s

At a glance: The STEP trials

A round-up of the STEP phase 3 clinical trials evaluating semaglutide for weight loss in people with overweight or obesity.
ADVANCED SEARCH
Log in
Top
European Journal of Epidemiology
Published in: European Journal of Epidemiology 5/2010

01-05-2010 | COMMENTARY

Don’t split your data

Authors: Henrik Källberg, Lars Alfredsson, Maria Feychting, Anders Ahlbom

Published in: European Journal of Epidemiology | Issue 5/2010

Login to get access

Abstract

False positive findings are a common problem in whole genome association studies. In this commentary we show that nothing is gained by randomly splitting a data sample to two equal sized subsets, where the first data subset is used for explorative purposes and the other sub set is used to confirm the findings in the first subset. We compare the random splitting procedure to using the full data sample for analysis, by using a Bayesian perspective with consideration taken to prior probability of a false positive finding.
Literature
1.
Thomas DC, Siemiatycki J, Dewar R, Robins J, Goldberg M, Armstrong BG. The problem of multiple inference in studies designed to generate hypothesis. Am J Epidemiol. 1985;122:1080–95.PubMed
2.
Satagopan JM, Verbel DA, Venkatraman ES, Offit KE, Begg CB. Two-stage designs for gene-disease association studies. Biometrics. 2002;58(1):163–70.CrossRefPubMed
3.
Thomas CD, Cassey G, Conti DV, Haile RW, Lewinger JP, Stram DO. Methodological issues in multistage genome-wide association studies. Stat Sci. 2009 (in press).
4.
Wacholder S, Chanock S, Garcia-Closas M, El Ghormli L, Rothman N. Assessing the probability that a positive report is false: an approach for molecular epidemiology studies. J Natl Cancer Inst. 2004;96(6):434–42.PubMedCrossRef
Metadata
Title
Don’t split your data
Authors
Henrik Källberg
Lars Alfredsson
Maria Feychting
Anders Ahlbom
Publication date
01-05-2010
Publisher
Springer Netherlands
Published in
European Journal of Epidemiology / Issue 5/2010
Print ISSN: 0393-2990
Electronic ISSN: 1573-7284
DOI
https://doi.org/10.1007/s10654-010-9447-3

Other articles of this Issue 5/2010

European Journal of Epidemiology 5/2010 Go to the issue

METHODS

A successful implementation of e-epidemiology: the Danish pregnancy planning study ‘Snart-Gravid’

PERINATAL EPIDEMIOLOGY

Selection by socioeconomic factors into the Danish National Birth Cohort

Methods

Comparison between web-based and paper versions of a self-administered anthropometric questionnaire

Erratum

Erratum to: Ignoring critique, attacking the critic

DIABETES MELLITUS

Short-term weight change and fluctuation as risk factors for type 2 diabetes in Finnish male smokers

Neuro-Epidemiology

Incidence of Creutzfeldt-Jakob disease in Taiwan: a prospective 10-year surveillance