Early thoughts on computing workshops for medical students

jp, 25 October 2017

I’m planning on running computing skills workshops for medical students at Queen’s University, and in this post I’ll share some of my thinking and questions on what this series ought to entail.

The gist of my workshop pitch is to equip future physicians with the computing literacy they’ll need to be researchers and research “consumers”. Our curriculum already contains basic instruction on study design, statistics, and proposal writing, but offers nothing about managing experimental data, cleaning it up, actually doing the statistics, creating useful figures and tables, or doing reproducible research. I’d like to change that.  My goal here is to introduce students to the tools and ideas of the trade, so that when they need them later on in their careers they know what is possible and how to learn them: I’d be so pleased at the end of the year if students could load up a dataset into Rstudio, and make a plot with a subset of the data on two variables with a linear fit line, while knowing and being confident enough to read through documentation to extend their work.

There is some prior art in this area of teaching that I’m aware of (please alert me if know of others!):

There are certain aspects of teaching medical students that makes my workshop series not necessarily a simple recapitulation of a Software/Data Carpentry/CAMH workshops:

  • Medical students have a dramatic (perceived) poverty of time
  • They don't have an immediate problem to apply these skills to (e.g. a thesis) which could serve as motivation to learn
  • My ignorance: what computing skills are relevant to clinicians?  In the same way Library Carpentry adapted Software/Data Carpentry teaching for librarians, I suspect I will need to do the same kind of translation. I haven't looked into any of the literature.

I feel as if I need to give some consideration to how I pitch these workshops to students, and how I run them so as to keep them interesting and relevant. My current plan is to run 2-3 hour sessions each month and keep them light, but focused and self-directed (we get enough lecture as it is) where I act as guide and perhaps interrupt to explain difficult concepts or run short interactive instructional pieces. I could set students off on lessons from Data/Software Carpentry, datacamp.com, R for Data Science, or the CAMH workshops. Speaking with Greg Wilson (founder of Software Carpentry, currently at datacamp.com), we also discussed teaching computer security (see cryptoparty.is) as one way to teach something immediately relevant to health care professionals (note: no one in my class I’ve spoken to about this seems to think this is at all interesting).

My next steps are to:

  • Invite students out to a brief meet up where I can pitch the series, answer questions, and get a feel for level of interest. I'm going to include an "ask me anything"/"explain like I'm five" session where I answer people's questions on computers and computing (I've found this to be a great way to warm people up).
  • Explore datacamp.com's offerings
  • Plan out the workshops for the year
  • Explore getting these workshops approved as an Academic Enrichment Program at Queen's, which would give students something to put on their medschool CV