The Best CS Summer Camp Paper: Sustainable, Effective, and Replicable
I’ve got a lot out of SIGCSE 2012, and I have several posts that I’d like to share. But I became ill on the last day of the conference, and am just now recovering.
I really do mean what I wrote in the title. I am, of course, biased towards the paper by my wife, Barbara Ericson, and our external evaluator on Georgia Computes, Tom McKlin, but I still think that this is the best paper on computing summer camps yet published at SIGCSE.
There are lots of people creating computing summer camps these days, and for good reason. They really can work for increasing student interest and providing some real education about computer science, which is missing from most U.S. schools. What makes Barb’s summer camp program so good that it really works on several levels:
- First, it is effective. They have reliable measures that students improve their attitudes about computing in pre/post comparisons. Women and members of under-represented groups in particular improve their attitudes about further study in computing. But even better: The students learn something about computer science. Barb and Tom have measures of learning about computer science and programming that indicate that the students in the summer camps are learning, too.
- Second, they are sustainable. Barb has created a business plan that makes these camps work continuously after only a $5K seed grant. Barb has been doing this for a long time, and she’s figured out several rules of thumb. For example, don’t have University faculty teach your camps. Faculty are too expensive, and high school teachers need and want the summer work — and it gives them the chance to learn something new to take into their classroom. Another example: Always offer both high school and middle school camps. High school camps give you the best chance to recruit undergrads into your program, but middle school camps can charge more (since the kids are too young to stay at home) and help cover the cost of the high school camp.
- Third, they are replicable. Through Georgia Computes, Barb has now given seed grants to start 11 more camps around Georgia. Some of these have been running for several years now. Better yet — they’re effective, too. The paper shows that the seed grant camps are returning results comparable to Barb’s original camps.
One of the things that I like best about Barb’s camps (besides the fact that they work) is that they benefit multiple levels of the computing education pathway. Barb offers workshops on “How to Run a Summer Camp” to higher-education faculty in Georgia, on logistics, curricula, and business plans. The faculty can apply for seed grants, keeping them involved. To get a seed grant, they have to show that they will have a sustainable business plan, that they will gather data for the evaluation effort, and that they will do something useful during the academic year with any robots or other kits that they purchase with the seed funds. We encourage faculty to set up “Lending Libraries” where the robots are made available to local teachers to use in their classes. The faculty then hire high school teachers, which gives them a chance to learn something new. Finally, the students get the camps.
It’s the combination of sustainable, effective, and replicable that really makes this a striking result. Summer camps can really work, and here’s a good paper on how. Sure, summer camps could be done even better, but I think that Barb has the current state-of-the-art.