Meta-research and Innovation in how research is conducted.
Meta-Research in the area of Methods is broad, aiming to develop, identify and promote best scientific practices while also quantifying the effect of questionable practices which lead to misleading or incorrect findings or interpretations. This research can include:
- Documentation of the prevalence of error, bias and questionable practices in study design, data collection, analysis, and interpretation, with empirical demonstration of their effect
- Uses, abuses, and misunderstanding of methods in statistical analysis, research synthesis and citation
- How to measure and control for “meta-biases;” differences in results seen among collections of studies with certain properties, but whose effect cannot be observed by looking at individual studies. These can include design (e.g. single vs. multi-center studies), sponsorship (private vs. government), country of origin, analytic approach, etc.
Innovation can help expand the use of optimal methods, for example by developing new approaches, methods, tools and policies to:
- Detect and correct for biases in meta-analyses or evidence evaluation
- Prevent bias and questionable research practices in study design, data collection, analysis, and interpretation
- Promote statistical literacy and merge frequentist methods with Bayesian thinking
- Optimize study design, maximize statistical power and information value, and prioritize research questions, for example to guide funding decisions in replication research
- Develop and promote tools to facilitate open science
- Promote scientific integrity, ethics and best research practices
Affiliated Centers & Programs:
CASB Group aims to increase the validity and credibility of scientific research...
Stanford Center is dedicated to advancing open and reproducible science...
CEDAR is an initiative of the National Institutes of Health to improve metadata...
Recent PublicationsAll Publications
Evaluation of Evidence of Statistical Support and Corroboration of Subgroup Claims in Randomized Clinical Trials
John Ioannidis , Joshua Wallach, Patrick Sullivan, John F. Trepanowski, Kristin L. Sainani, Ewout W. Steyerberg Methods
- Journal Title: JAMA Intern Med
- Publication Date: 2017
- Issue #: 4
- Page #: 554 - 560
Steven Goodman , Sebastian Schneeweiss, Michael Baiocchi Methods
- Journal Title: JAMA
- Publication Date: 2017
- Issue #: 7
- Page #: 705 - 707
Pertile S.D.L., Moreira V.P., Rosso P. Methods
- Journal Title: Journal of the Association for Information Science and Technology
- Publication Date: 2016
- Issue #: 10
- Page #: 2511 - 2526