Teaching basic lab skills
for research computing

2015 Post-Workshop Instructor Debriefing, Round 7

The mentorship team held our latest round of post-workshop debriefing sessions for instructors who taught recently. Instructors from workshops at Clemson University, University of Miami, University of Melbourne, University of Oklahoma Libraries, Harvard School of Public Health, and Weill Cornell Medical College joined us for our discussions, as well as a few new instructors who will be teaching in coming weeks. Here's a recap of common themes and highlights:

Things that worked well:
  • Teaching R and Python together: University of Melbourne ambitiously covered Unix, Git, SQL, Python and R over a three day workshop. Although they reported some learning fatigue near the end, the lesson pacing was appropriate and they received largely positive feedback. R and RStudio were taught first, followed by Python and a short introduction to Jupyter Notebook, which allowed instructors to highlight parallels between the scripting language and their respective interfaces.
  • Making Git interactive: Git can be a difficult topic for beginners to comprehend, but there are a few methods our instructors reported working well. Building a concept map during the lesson can help students build a mental framework for how the parts of Git work together (example picture from Weill Cornell Medical College). Second, pairing students up to work on the collaboration with GitHub lesson has been received well in multiple workshops.
  • Word clouds to summarize student feedback: Instructors from University of Melbourne compiled student feedback each day and presented wordclouds using previous methods taught (R, Python, shell, repo here). This seems like a great way to showcase how to perform the same task in different ways.

Specific challenges:

  • Pre-workshop preparation: Multiple instructors noted issues while preparing for their workshops. These are especially important concerns because 1) we want to encourage "repeat customers," and 2) there may be altered expectations from organizers given the recently instituted workshop fee. In particular, assigning an experienced lead instructor (from the group of participating instructors) well in advance is essential. This can help ease communication between instructors and local organizers on the details of planning (logistics, presenting appropriate material for the audience, finding helpers). These planning steps are important for making sure students receive the pre-workshop survey and instructions for installing software prior to arriving, as installing software on-site at the beginning of the first lesson is time consuming and can be impeded by lackluster internet. If students are unprepared from the start, point them to instructions so they can install subsequent software during lunch or breaks. Alternatively, build time into the schedule to allow for installation.
  • Crafting material to particular audiences: Several instructors noted difficulty in meeting the particular skill levels of their students. In one case, the R material seemed too advanced, so instructors substituted modified Data Carpentry material. On the other hand, challenge questions for shell/Git/Python appeared too easy for students in other workshops. Several lessons were not completed because of time constraints; cutting lesson material to fit the workshop is a continuing problem for instructors. However, multiple instructors noted the usefulness of domain-specific datasets, and the University of Oklahoma group supplemented their lessons on Unix, Python, and Git with specific discussions of library data (MARC).
  • Displaying materials for instructors/students: Instructors are often compelled to juggle multiple windows when teaching, including the website for materials, shell/console for entering commands, Etherpad, and/or slides. Use of dual projectors or a split display, especially to model collaboration in GitHub, can help ease this pressure. We also discussed strategies to help instructors prepare so that they don't need to reference teaching materials on the projected screen (printing off materials, using a tablet/additional laptop, making a short list of teaching notes that include commands to cover).


  • Differentiating between Software and Data Carpentry: Many students are asking instructors about distinctions between the two Carpentry groups. Although the nuances at this point are largely unimportant to very new learners, switching out some lessons (see Challenges, above) can lead to some confusion on the part of students. It would be useful to have clear language that relates these two groups as they continue to develop their own personalities.
  • Dialogue & Discussion

    You can review our commenting policy here.