Unit of analysis issues in laboratory-based research

Many studies in the biomedical research literature report analyses that fail to recognise important data dependencies from multilevel or complex experimental designs. Statistical inferences resulting from such analyses are unlikely to be valid and are often potentially highly misleading. Failure to...

Full description

Saved in:
Bibliographic Details
Published ineLife Vol. 7
Main Authors Parsons, Nick R, Teare, M Dawn, Sitch, Alice J
Format Journal Article
LanguageEnglish
Published England eLife Sciences Publications Ltd 10.01.2018
eLife Sciences Publications, Ltd
Subjects
Biomedical research
Clinical trials
Epidemiology and Global Health
experimental design
Experiments
Generalized linear models
Human subjects
Laboratories
Medical research
mixed-effects models
Public health
Science
Science Forum
Scientists
Statistical analysis
Statistical methods
Statistics
United Kingdom > UK
global health
epidemiology
experimental design
Science Forum
mixed-effects models
statistics
Online AccessGet full text

Cover

More Information
Summary:Many studies in the biomedical research literature report analyses that fail to recognise important data dependencies from multilevel or complex experimental designs. Statistical inferences resulting from such analyses are unlikely to be valid and are often potentially highly misleading. Failure to recognise this as a problem is often referred to in the statistical literature as a unit of analysis (UoA) issue. Here, by analysing two example datasets in a simulation study, we demonstrate the impact of UoA issues on study efficiency and estimation bias, and highlight where errors in analysis can occur. We also provide code (written in R) as a resource to help researchers undertake their own statistical analyses.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
These authors contributed equally to this work.
ISSN:2050-084X
2050-084X
DOI:10.7554/eLife.32486