class: center, middle, inverse, title-slide # Shifting Science and Research Culture Towards Openness and Reproducibility Through Education and Training ### Luke W. Johnston --- layout: true <style type="text/css"> .footer-left { background-color: #FFFFFF; position: absolute; bottom: 8px; left: 20px; height: 60px; width: 30%; font-size: 14pt; } .footer-right { background-color: #FFFFFF; position: absolute; bottom: 8px; right: 5px; height: 50px; width: 30%; font-size: 14pt; } </style> .footer-left[
] .footer-right[ [slides.lwjohnst.com/misc/2021-11-18](https://slides.lwjohnst.com/misc/2021-11-18/) ] ---
# Outline <!-- 10 min --> - Key points - Introduction to course and attendees - Key principles used while creating/updating course - Lessons learned - Experiences of effectiveness and impact (within Denmark) - Summary ??? Hi everyone, thanks for coming to listen to my talk. I'm Luke and I work as a postdoc at the Steno Diabetes Center in Aarhus Denmark. If you want to check out these slides, there's a link to them in the footer of each slide. So, since this is an audience of other trainers, I'm going to focus on the topics of practicals of teaching and on curriculum development. (Go over outline). While talking about key principles and lessons learned, it will tie back to some of the key messages. --- # *Actionable* key 🔑 points ??? Before I begin, I want to say a couple of caveats about these key messages. Here, I'm assuming that the learners are working/full-time researchers (including PhD students), not necessarily in an undergraduate context and related to learning data analysis or more practical type skills. -- 1. Have material be online, easily found - Follow it (almost) exactly, deviate only if necessary 1. Encourage participants to instruct future workshops - Provide onboarding documentation and support 1. Emphasize code of conduct and safety of the learning space - *Embody* it 1. Use and weave in reading, listening, doing, and discussing activities *in class* ??? (Go over the messages). --- class: middle # Reproducible Research in R (R3 or r-cubed) course/workshop for PhD students and postdocs *doing biomedical research* .footnote[ - Website: [r-cubed.rostools.org](https://r-cubed.rostools.org) - JOSE paper: [10.21105/jose.00122](https://jose.theoj.org/papers/10.21105/jose.00122) - Also intermediate course: [r-cubed-intermediate.rostools.org](https://r-cubed-intermediate.rostools.org/) ] ??? Now getting into the course itself. In Denmark, a course is anything from a couple days to a couple weeks, while in North America this might be called a workshop. The course is teaching reproducible research in R to PhD students and postdocs who do biomedical research, largely diabetes research. This course is 3 full days, composing of 5 code along sessions where the instructor types and the learners follow along, a few lectures, and a final group project. For more info on the course, check out the links below. --- class: middle # Course arose from need for skills in data analysis 👩🏽💻 ## ... and for more awareness on reproducibility and open science 🤯 ??? While I've been teaching these general topics since my Masters, this course specifically I started during my postdoc because one, there was a need for more computational skills in my field and two, because the awareness around reproducibility and open science was very lacking. --- class: middle # (Some) Key 🔑 principles when developing the material and learned from the feedback ??? When developing this course along with updating each time we run it, there were some key principles that we used to form the foundation for the entire content of the course. And I'll be covering some of them for most of the talk. --- ## Use evidence based learning and teaching best practices 👩🏻🏫 ??? The first one is, of course, building the curriculum based on evidence around learning and on teaching best practices. -- 1. Collect (and use) anonymous and critical feedback often! Including from instructors ??? ### 1 Like any research, in order to build and then improve something you need to actively get feedback, often. I made a Google Forms survey that asks generally, what worked and what could be improved, at the end of each day. Then I use that to fix, remove, or keep things. -- 2. Multiple activities to learning *in class* (reading, doing, listening, discussing, teaching, group, and solo) ??? ### 2 Learning requires multiple different angles repeating basically the same information in a different way until it gets remembered and saved into long-term memory. In this way, during the course we get learners to do multiple tasks like reading activities, hands-on exercises, listening, and so on. In relation to reading, one thing I found was talking about a concept always had some variance in saying it each time I ran the course or when another instructor did the specific session. Plus, we have many people who have English as a second language, so the amount of information you could convey orally was much less than by reading. It also slows us down, because, I do it, its very easy to go too fast through the material. After they read a section, as the instructor we would briefly go over the topic again, to reinforce the concept. After adding the reading activities and getting the feedback, there was consistent and strong positive comments about that, so we kept it and added more. -- 3. Strong and focused narrative tied to learning outcome: Less is more ??? ### 3 The narrative throughout the whole curriculum should be laser focused on the learning outcome. Brains can only hold so much information in them at any one time, so the key thing while building material is keep things lean and focused on what the learning outcome and objectives. Each time we've ran the course, we've stripped out more and more. Don't deviate from the material. Plus, everytime I deviate too much from the material because I have some expert knowledge that I want to share from excitedness, I see it in the feedback that it confused them or that if I talk about it it should be in the material. -- 4. Limit lectures and slides (largely used for conceptual/ranting topics) ??? ### 4 Lastly, lectures are some of the poorest form of memory retention, so we severely limit them to only very abstract topics or things to rant about, like how un-open science is. --- class: middle ## Openly licensed
and easily accessible
??? The next key principle is to ensure the material is maximally used and re-used. This means applying an open license to it and having it easily accessible online. I think this is a bit of a misunderstood area. I see a lot of for instance university courses that don't have their material publicly available, thinking that they need to keep it only for people paying or attending the class. While I understand that idea, I think that's a shame and maybe misguided because people don't usually take courses for the content itself, generally, they take them to be taught. They come for teachers and the experience, not the content alone. And we see it in the feedback, there is always very strong and consistent positive feedback talking about how great it is that the material is online and easily accessible. --- class: middle ## Safe
and supportive
environment .footnote[Through a code of conduct, explicit expectations of participants and instructors, as well as detailed and descriptive syllabus.] ??? This next principle is something I think is so important... and that is establish and maintaining an environment that is safe and supportive for learning. Have the explicit code of conduct on expected behaviour of everyone involved at the forefront and have instructors embody this attitude. Learning is so dependent on the learners motivation and interest and engagement with the content... and if they don't feel safe, they won't learn as much as they could. And it's so easy to unconsciously use dismissive language like saying "simply" or hand-waving a topic away when you're teaching. But it can lead to undermining learners confidence and trust in themselves, particularly for underrepresented or minority groups within science. So constantly self-reflecting is really vital to this topic. --- class: middle ## Design material for participants *as well as* potential/new instructors .footnote[As well as potential contributors (i.e. the participants).] ??? Lastly, the material isn't only for the participants, it's also for future instructors and contributors. Since it is written with the idea that learners will read it during the course or afterwards, this also gives new instructors something to use to prepare for teaching their session or course itself if they want. Interspersed throughout the text are messages to the instructors on top of an instructor guide giving advice and tips on how to prepare, what to do, and how to teach. We hope that, structured as it is, anyone can take the course and try to teach it themselves to others. --- class: middle # Experiences of effectiveness and impact (within Denmark 🇩🇰) .footnote[Feedback data is found on the [GitLab repository](https://gitlab.com/rostools/r-cubed/-/tree/main/feedback).] ??? Alright, I've gone over the key principles for the course, how effective and impactful has it been? We've ran the course through the Danish Diabetes Academy here in Denmark, which is an organization that provides funding, networking events, and educational events and courses to diabetes researchers. They collect evaluations for the course and both this course as well as the intermediate one are both the highest rated and most in-demand courses that they offer. Plus based on the feedback we collect, we can see how much the participants really enjoy the course and often recommend it their colleagues. If you want to read some of the feedback we get, they are found in our GitLab repository, the link is there in the footnote of this slide. --- # Summary and recap 1. Have material be online, easily found - Follow it (almost) exactly, deviate only if necessary 1. Encourage participants to instruct future workshops - Provide onboarding documentation and support 1. Emphasize code of conduct and the safety of the space to learn - *Embody* it 1. Use and weave in reading, listening, doing, and discussing *in class* ??? Anyway, I could talk about this for a long time. But to end, let's go back to the key messages. I talked about (go over messages). -- ### Try out the material and give us feedback on it! 🤓 ??? And to end, please, if you try out the material, lets us know! We'd love more feedback on it! Thanks for listening! # Appendix Why not use existing material from e.g. Carpentries? - No strong narrative thread - Jumping around with concepts without reusing them again later - Using older workflows/packages - Emphasis/focus on writing software - Not targeted to biomedical researchers