What are the best ways to ensure research collaborators use appropriate statistical methods?

1 Establish clear goals and expectations

Before you start any collaborative research project, make sure you have a clear and shared understanding of the goals, hypotheses, and research questions that guide your work. Discuss how you will collect, manage, and analyze your data, and what statistical methods and assumptions are suitable for your data type and design. Agree on the level of significance, confidence intervals, and effect sizes that you will report and interpret. Also, clarify the roles and responsibilities of each collaborator, and how you will communicate and document your progress and results.

2 Choose appropriate tools and platforms

Using the right tools and platforms can make your collaborative research easier and more efficient. For example, you can use online platforms such as Google Drive, Dropbox, or GitHub to store, share, and synchronize your data and code files. You can also use software such as R, Python, or SPSS to perform your statistical analyses, and tools such as R Markdown, Jupyter Notebook, or Knitr to create reproducible reports that combine code, output, and narrative. These tools and platforms can help you ensure consistency, transparency, and quality of your statistical methods and results.

3 Seek feedback and support

Collaborative research is an opportunity to learn from each other and improve your skills. Don't hesitate to ask for feedback and support from your collaborators, especially if you are not familiar with certain statistical methods or concepts. You can also consult external experts, such as statisticians, data scientists, or methodologists, who can provide guidance and advice on your data analysis and interpretation. Additionally, you can use online resources, such as blogs, podcasts, books, or courses, to learn more about statistical methods and best practices.

4 Validate and verify your results

One of the most important steps in collaborative research is to validate and verify your results. This means checking that your data are accurate, complete, and reliable, that your code is error-free and well-commented, that your output is consistent and reproducible, and that your interpretation is valid and sound. You can use various techniques, such as data cleaning, debugging, testing, cross-validation, sensitivity analysis, or peer review, to validate and verify your results. You can also compare and contrast your results with those of your collaborators, and discuss any discrepancies or disagreements.

5 Share and disseminate your findings

Finally, you want to share and disseminate your findings with your collaborators, stakeholders, and the wider scientific community. This can include preparing and presenting reports, papers, posters, or slides that summarize your data analysis and results. When you share and disseminate your findings, make sure you follow the ethical and professional standards of your field and journal, such as citing your sources, acknowledging your contributions, disclosing any conflicts of interest, or sharing your data and code. You should also be prepared to answer questions, address comments, and receive feedback from your audience.

6 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?