In March I coordinated a Software Carpentry workshop here at the University of Arkansas. We did a two-day workshop on R, the Unix Shell and Git and it was very well received. A few people have asked me to write up my thoughts on planning the workshop, the workshop itself, etc, so here it goes.

I decided to try and have this workshop at UA because there was a need within the graduate students in our department to learn how to use R and how to do more reproducible science, but it wasn’t being taught in any of our courses and not very many of our faculty are heavy R users, so most of them were coming to me or a few other grad students for help. Which was fine, I love helping people, but wasn’t super efficient.

I learned about SWC through twitter and looked into what was required to have a workshop. I threw together a budget of the costs (hotel, flights and meals for instructors as well as the fee charged by SWC) and then approached my department head to see if the department would be willing to fund the workshop. He agreed to part of it and i began contacting other department heads, all of which turned me down for various reasons. Then my department head suggested I contact the graduate school and through a combination of the graduate school and the department of biological sciences we were able to get the funding together.

From there I contacted SWC, and a short time later they had lined up two instructors for us and then it was lots of paperwork to get plane tickets bought and hotel rooms reserved and we were on our way.

The unexpectedly hard part of this (which in hindsight I should have seen as a big road block) was finding a room. We had this workshop on a Thursday/Friday right before spring break, so classes were still in session, so finding a room we could take over for two entire days, was very very challenging. I went around to 6 different offices before I finally found this fairly awful room that we could have access to. I reserved it and was later that day talking to our wonderful administrative assistant when it occurred to us that I could use the student lounge here in the department. So I canceled the awful room and went about finding tables and chairs to turn our small undergrad student lounge into a ‘classroom’ of sorts.

I did not charge anyone for the workshop, the moving around of money here to do so was going to be very messy and so I decided to deal with the no-shows and not charge anyone. We had room for 30 people (had we had a larger room I would have done 40, but space was limited). I registered 32 figuring we’d have no shows, and it worked out perfect, 30 people attended, mostly graduate students from five different departments, but also one undergraduate and two faculty.

We were able to buy some snack stuff for breaks via the money allocated to us by the department/grad school. I opted to go for the healthy food route, so fruit, yogurt, granola bars and veggies in addition to a few donuts. Everyone really seemed to like this. We also had abundance coffee and hot water for tea. One thing I wish I had done different was requesting everyone bring their own coffee mug, I HATE Styrofoam, but was forced to use it, because of my own poor planning.

Advertising wise I had emails sent out through the biology graduate student assocation and our department ‘list serve’ (list of emails our administrative assistant has and forwards things to). From there it got spread around to other people outside the department. My original plan was to advertise many other places, but we were almost totally full 24 hours after the first email went out, so I didnt have to.

I did not advertise the workshop as being about statistics, I used very similar language to what is on SWCs website, but we still had two participants who expressed some disappointment since we did not teach them how to do GLMs and Anovas in R. I’m not sure how to prevent this from happening again other then adding text saying ‘this is not a statistics workshop’ to fliers. I think what may have lead to this for these two participants was not understanding that R can do more then stats, which is a totally understandable misconception if you know very little about R.

One person suggested we make the two days of the workshop separate, in that you can sign up for one or the other, since they didn’t like the UNIX/Git stuff, and I appreciate that not everyone will be super pumped about learning those skill sets, but most people in the workshop said they really liked the material covered, they all just wished we had more time.

One challenge we faced was most students in our department are on teaching assistantships, and so many had to duck in and out to go teach labs, in the future I would try to schedule this over a break, or try and make some kind of arrangement for TAs, as many of them were very sad to have to leave and come back and miss material.

Overall it was a huge success, many are asking me to arrange another workshop, but two of our faculty (one of which attended the workshop) are going to be teaching a course this fall which will mirror much of the material covered in this workshop, so I’m not sure how soon we will need another one. But I am glad we were able to have it. As I continue to help my peers with R I am really amazed with how much their skills have improved and their ability to solve their own issues now, which is really fantastic to see.

In summary, awesome experience, would do again, if you have specific questions about this workshop and my experience feel free to drop me a line (Twitter - @RallidaeRule or email aurielfournier@gmail.com)