Archives

Make your research reproducible – a hands-on course (2024)

Do you know that feeling, when you work on your research project after not having looked at it for a while (for example after receiving reviews) and you can’t remember how the different parts fit together?

And when you finally figure it out again, the results are somehow different. Oh no!

In this course we will set you up to never run into these kinds of issues again.

Let’s get your research project organized, under version control, stable, and published so that you can confidently say that your research is reproducible.

Topics

  • The goal of producing FAIR and reproducible research outputs
  • Working on research projects as a team
  • Get your project organized: organized folders, naming, documentation
  • Set up reproducible workflows: version control with git, stable computing environments, and automation
  • Make your project reproducible and FAIR for all: publication platforms, licensing, and more

Format

This is an online course. Each week you will:

  • watch the 📺weekly videos (~45 min),
  • review one of the 📖 booklet chapters,
  • have a short session with your 👤 accountability buddy* (15-20 minutes), 
  • implement the ✍️ tasks of the week, and
  • discuss your progress in the 👩‍🏫 weekly online meeting with the instructor and fellow course participants (1-1.5 hours).

* You will choose your accountability buddy during the course. You and your buddy will help each other in implementing the tasks of each week.

Weekly Meetings

The course includes 5 Online Meetings, in which you will discuss the week’s contents with the instructor and fellow participants:

Meeting 1: Sep 25, 3:00 – 4:30 pm CEST
Meeting 2: Oct 02, 3:00 – 4:30 pm CEST
Meeting 3: Oct 09, 3:00 – 4:30 pm CEST
Meeting 4: Oct 16, 3:00 – 4:30 pm CEST
Meeting 5: Oct 23, 3:00 – 4:30 pm CEST

Prerequisites

  • Basic programming knowledge (R, python, …)
  • Willingness to learn new technical skills

About the Instructor

Heidi Seibold holds a PhD in Biostatistics with a focus on Machine Learning. Her research has been in the intersection of Data Science, Reproducibility and Medicine. She is a solopreneur in Open and Reproducible Research and an independent researcher at IGDORE. Heidi is the co-founder of the Digital Research Academy and Open Science Freelancers.

StatTag and StatWrap for Conducting Collaborative Reproducible Research

Challenges of reproducible research

Practicing reproducible research is important, but increasingly complex as studies involve more data and statistical code, and larger teams. Adopting reproducible research workflows can be especially daunting for research teams with a diverse set of needs, skills, and expectations for software tools.

For example, in medical research, most manuscripts are prepared in Microsoft Word, leaving clinicians to copy and paste, or even re-type, statistical estimates into Word documents. In contrast, statisticians may use R Markdown or Jupyter Notebook to generate reports weaving together statistical results with interpretation, but their collaborators may be unwilling to draft manuscripts in these programs. In addition, teams may struggle to communicate and keep track of information such as: Who worked on the analyses, when, and what decisions did they make? Where is the most recent data? What are the code file dependencies and code libraries?

Get to know StatTag and StatWrap

This talk will describe two software tools designed to address these problems — StatTag and StatWrap — both of which grew out of the challenges of conducting collaborative research in an academic health center. StatTag addresses a need to integrate document preparation in Microsoft Word with statistical code and results from R, Stata, SAS, or Python. StatWrap is an assistive, non-invasive discovery and inventory tool to document the evolution of project, combining automatically collected metadata (e.g., statistical packages, code file dependencies), investigator-supplied documentation (e.g., analysis notes, personnel), and source control. Both StatTag and StatWrap are free, open-source software programs designed to promote the conduct of reproducible research, especially for collaborative teams.

This event takes place on Thursday, September 28 at 10.15am – 11.15am as part of the colloquium of the Department of Statistics of the LMU Munich. We invite everyone interested to register for joining the colloquium online.

About the instructor

Dr. Leah J. Welty’s research interests include the application of biostatistics to psychiatry and environmental research and the development of software tools for reproducible research. She leads the development team for StatTag (stattag.org) – free, open-source software connecting Microsoft Word to R, SAS and Stata. She is also the lead biostatistician for the Northwestern Juvenile Project, a large-scale longitudinal study of psychiatric disorders and risky behaviors in delinquent youth, as well as NJP: NextGen, a study of the children of the original Northwestern Juvenile Project participants. She directs the Biostatistics Collaboration Center, Feinberg’s core biostatistics resource for non-cancer research. (Northwestern University 2023)

Make your Research Reproducible – A Hands-On Course

Do you know that feeling, when you work on your research project after not having looked at it for a while (for example after receiving reviews) and you can’t remember how the different parts fit together? And when you finally figure it out again, the results are somehow different. Oh no!

In this course we will set you up to never run into these kinds of issues again. Let’s get your research project organized, under version control, stable, and published so that you can confidently say that your research is reproducible.

Topics

  • The goal of producing FAIR and reproducible research outputs
  • Working on research projects as a team
  • Get your project organized: organized folders, naming, documentation
  • Set up reproducible workflows: version control with git, stable computing environments, and automation
  • Make your project reproducible and FAIR for all: publication platforms, licensing, and more

Format

This is an online course. Each week you will:

  • watch the weekly videos (~45 min),
  • review one of the booklet chapters,
  • have a short session with your accountability buddy* (15-20 minutes), 
  • implement the tasks of the week, and
  • discuss your progress in the weekly online meeting with the instructor and fellow course participants (1-1.5 hours).

* You will choose your accountability buddy during the course. You and your buddy will help each other in implementing the tasks of each week.

Weekly Meetings:

  • Tuesday, October 17, 3 pm – 4:30 pm CET
  • Tuesday, October 24, 3 pm – 4:30 pm CET
  • Tuesday, October 31, NO MEETING
  • Tuesday, November 7, 3 pm – 4:30 pm CET
  • Tuesday, November 14, 3 pm – 4:30 pm CET

Prerequisites

  • Basic programming knowledge (R, python, …)
  • Willingness to learn new technical skills

About the Instructor

Heidi Seibold holds a PhD in Biostatistics with a focus on Machine Learning. Her research has been in the intersection of Data Science, Reproducibility and Medicine. She is a solopreneur in Open and Reproducible Research and an independent researcher at IGDORE. Heidi is the co-founder of the Digital Research Academy and Open Science Freelancers.