How to report statistics in journal articles: Guidelines and best practices
How to report statistics in journal articles: Guidelines and best practices
Author Resources > Writing a methods section > Reporting statistics in journal articles
A statistical analysis is a key component of the methods and results sections of a manuscript. Along with clear methods and transparent study design, the appropriate use of statistical methods and analysis impacts editorial evaluation and readers’ understanding of the study. Learn how to report statistical analyses for publication success and improve future reproducibility.
Rules for reporting statistical analysis
The statistical methods employed in research must always be:
- Appropriate for the study design
- Rigorously reported in sufficient details for others to reproduce the analysis
- Free of manipulation and selective reporting
Tips for reporting statistics
Define your analytical methodology before you begin.
Consider and develop a thorough study design that defines your line of inquiry, what you plan to do, what data you will collect, and how you will analyze it. (If you applied for research grants or ethical approval, you probably already have a plan in hand) Revisit your study design throughout the research—and stick to it.
Preregister your study design with a journal.
Many journals offer preregistration, allowing peer review of your study design before research begins. If approved, you receive provisional acceptance, boosting credibility and reducing bias. Declaring your analysis plan early strengthens your research and improves publication chances—even with null results.
Imagine replicating or extending your own work years in the future.
What would you need to know to replicate or extend your own work? Consider:
- Which details would you need to be reminded of?
- What did you do to the raw data before analysis?
- Did the purpose of the analysis change before or during the experiments?
- What participants did you decide to exclude?
- What process did you adjust during your work?
Even if a necessary adjustment you made was not ideal, transparency is the key to ensuring this is not regarded as an issue in the future. It is far better to transparently convey any non-optimal techniques or constraints than to conceal them, which could result in reproducibility or ethical issues in the future.
Checklists and guidelines for reporting statistics in specific disciplines
You can apply the open science practices outlined above no matter what your area of expertise—but in many cases, you may need more detailed guidance specific to your own field. Many disciplines, fields, and projects have developed guidelines and resources to help with statistical reporting. Some available checklists for specific disciplines include:
Biomedical Research
SAMPL guidelines
The “Statistical Analyses and Methods in the Published Literature” (SAMPL) guidelines covers basic statistical reporting for research in biomedical journals.
General
PLOS One guidelines for statistical reporting
While specific to PLOS One, these guidelines should be applicable to most research contexts since the journal serves many research disciplines.
Systematic reviews & Meta analyses
PRISMA
The "Preferred Reporting Items for Systematic Reviews and Meta-Analyses" (PRISMA) is an evidence-based minimum set of items focusing on the reporting of reviews evaluating randomized trials and other types of research.
Life Sciences
MDAR checklist
The "Consistent reporting of Materials, Design, and Analysis" (MDAR) checklist was developed and tested by a cross-publisher group of editors and experts in order to establish and harmonize reporting standards in the Life Sciences. The checklist, which is available for use by authors to compile their methods, and editors/reviewers to check methods, establishes a minimum set of requirements in transparent reporting and is adaptable to any discipline within the Life Sciences, by covering a breadth of potentially relevant methodological items and considerations.
