Communities of practice

Many times there is nothing left but to be self-taught. I remember spending entire days in the university library binging books, reviewing topics that would be part of the midterms and final exams, but that were not covered in class. Although I love studying and learning on my own, it was frustrating to constantly doubt whether I was understanding correctly what I was reading, because I had no one to discuss it with. Looking back, I realize that I was more lonely and did not seek support in study groups. I think it was a very competitive environment and it was not easy to say among colleagues “I don’t understand”.

It was during my master’s and doctorate program that I found the communities of practice, friendly spaces that gather people with the same interest in learning, sharing or practicing a tool or a topic. These collective spaces have, in general, the commitment to keep the environment and conversations safe and friendly for all their attendees (aka “Code of Conduct”). Besides, in these spaces there is a unique horizontality, where regardless of whether you are a novice or an expert in a field, we all treat each other in the same way. And it is thanks to these spaces that I have learned relevant tools for my daily work, I met amazing and generous people from whom I learn a lot and who have opened doors that would not have been within my reach in other circumstances.

The Turing Way project illustration by Scriberia. Zenodo. http://doi.org/10.5281/zenodo.3332807

Communities and reproducibility in science

As an academic researcher in health sciences, I feel responsible for making my research reproducible. That is, if someone else follows every step of my statistical analysis, with the same data, they would obtain the same results that I obtained in my study. As obvious as this may sound, there is a huge reproducibility crisis in science internationally and covering different disciplines, so we should all do better. Also, if it weren’t for the authors publishing their code, implementing complex analysis would often be much more difficult. And finally, another reasons why it is worth doing reproducible work is to make your life easier in the future, when you have to re-analyze your work because the data was updated or because a reviewer asked you to modify the current analysis or perform a new analysis.

Dan Quintana: Five things about open and reproducible science that every early career researcher should know. doi: 10.17605/OSF.IO/DZTVQ https://osf.io/dztvq/

Since many of the topics are not covered in formal education, many communities of practice have generated very valuable content and many talks so that we can learn how to make our research reproducible. So in order to motivate those who read this post, I am going to summarize some of the communities in which I participate/organize.

R-Ladies

R-Ladies is a global organization that promotes gender diversity in the R user community. It has more than 130 chapters around the world and is in 56 countries (more stats here). In 2017, together with Laura Ación, we founded R-Ladies Buenos Aires (the story behind it is in this blog post: vacation <- grow(RLadiesNetwork) and since then I am a big fan of the community 💜. I actively contribute as an organizer in R-Ladies Rotterdam and collaborate in R-Ladies La Paz.

Learning R has a slower learning curve than other programs, but whether it is easier or more complicated depends a lot on the pedagogy with which it is taught. R-Ladies is generating a large number of materials, workshops and events to learn R in the most friendly way possible. Many of the tutorials for newcomers in R focus on learning the tidyverse and collective efforts are working on translating essential materials into Spanish such as the R for Data Science book and the package datos. All events are free and due to the pandemic they have been virtualized, so there are events all the time (see the available events here).

Beyond learning about R, I made wonderful friendships through R-Ladies and I found a place were I can share my passion for data science with brilliant women who work in totally different disciplines. Thanks to this community I understood that we need protected spaces to empower ourselves as women, and to lighten the burden of living in patriarchal societies.

Reprohack

Reprohack is a growing community, born as a hackathon where participants gather to reproduce published articles that have code and data available. At the end of the hackathon, participants give a feedback to the authors. You can participate in the hackathon in different ways:

• Presenting an article to be reproduced during the hackathon: The benefit of participating in this way is that you receive a friendly and positive feedback about the reproducibility your work, in addition to receiving an appreciation of the effort involved. It is also an opportunity to disseminate your research with a wider network. Great twitter thread on the value of sharing your paper for #reprohack here.
  
• Reproducing a paper during the hackathon: This allows you to have a hands-on experience on reproducibility and learn tools and ways of making reproducible work. In addition, complementary talks are usually given during these sessions.
  
• Organizing a reprohack: I had the opportunity to organize the first Reprohack in Leiden, the Netherlands and it was an amazing experience! We wrote a post about it here and wrote a paper about it here. There are many ways to organize a reprohack, depending on who is your audience. The Reprohack Github repo has all the info and materials so you can host your own events.

Thanks to Reprohack I was able to learn new tools (beyond R) and I met great people who are in a similar situation to mine (doing a PhD) and with whom I can share how difficult and fascinating is the world of reproducibility and open science.

Open Science Community Rotterdam (OSCR)

OSCR is a community managed by different academic institutions in Rotterdam, and it has different chapters in other cities of the Netherlands. This community aims to promote and adopt open and transparent scientific practices throughout the research cycle.

The community hosts free events about good practices of open science and collaborates with other communities, for example the Journal Clubs are a collaboration with the community of ReproducibiliTea and many seminars are the result of collaborative work with RIOTS - Reproducible, Interpretable, Open, Transparent - Science Club. In June 2020 I had the opportunity to give a workshop on tidyverse as an R-Studio certified instructor in an online event. All materials of the workshop are available here.

Other communities of practice

If you are interested in knowing more about other Communities of Practice in Latin America and in the world, you will be interested in the video recorded by more than 20 authors from Latin America (including myself), for the international conference of R users: useR!2020.

In this video and blog post we highlight the communities and organizations in Latin America in which we are involved and international communities who have community members from Latin America.

If you are interested in knowing more about these communities, please reach out!