Common statistical errors in systematic reviews: What to look for and how to spot them

Topic category Editorial processes and supporting review authors
Date and Location
Date: 
Wednesday 3 October 2012 - 13:30 - 15:00
Location: 
Methods Group
Methods Group: 
Statistical Methods Group
Contact persons and facilitators
Contact person Affiliation Country
Joanne McKenzie (Contact this person) School of Public Health and Preventive Medicine, Monash University Australia
Facilitator Affiliation Country
Joanne McKenzie School of Public Health and Preventive Medicine, Monash University Australia
Peter Herbison University of Otago New Zealand
Marialena Trivella UK Cochrane Centre United Kingdom
Elaine Beller Centre for Research in Evidence-Based Practice, Bond University Australia
Other contributors
First nameLast nameAffiliation and Country
First name: 
Steff
Last name: 
Lewis
Affiliation and Country: 
Centre for Population Health Sciences, University of Edinburgh, UK
First name: 
Helen
Last name: 
Worthington
Affiliation and Country: 
Cochrane Oral Health Group, School of Dentistry, University of Manchester, UK
First name: 
Chris
Last name: 
Cates
Affiliation and Country: 
Population Health, St George’s University of London, UK
Target audience
Target audience: 
Co-ordinating Editors and Managing Editors, review authors
Is your workshop restricted to a specific audience or open to all Colloquium participants?: 
Open
Level of knowledge required: 
Any
Type of workshop
Type of workshop: 
Discussion
Abstract text
Abstract: 
Objective:
To provide Co-ordinating and Managing Editors of Cochrane Review Groups and review authors with an overview of common statistical errors in systematic reviews, what to look for, and how to spot them.
 
Description:
In this workshop we will cover statistical and interpretational errors that commonly occur in systematic reviews. Topics covered will include issues surrounding outcomes, meta-analytic models (fixed effect versus random effects), non-standard designs, risk of bias, selective reporting, and interpretation of results. Suggestions of how potential errors may be detected will be provided along with examples. We will also discuss how some errors may be avoided through consideration of the issues when developing review protocols. The workshop will include a mix of taught content and discussion.