On Mon, Jun 12, 2017 at 3:26 PM, Amanda Holland-Minkley <[log in to unmask]> wrote:

As far as the larger question of, do we need a survey on these topics, I'm wondering what our end goal is as a group. I know that this is a short-term committee and that we need to assemble a report (this summer, I think?).

In some sense our end goal is defined by the Committee's charge from SIGCSE (see http://sigcse.org/sigcse/programs/committees/liberal), which is to confirm or refute two apparent needs,but  also to identify any other needs and suggest ways of meeting any or all of said needs. The two apparent needs are (1) for a "voice" that can speak for liberal arts computing programs to the larger computing community, e.g., on ACM curriculum committees, and (2) for a place -- perhaps a virtual one -- where members of the liberal arts CS community can get together with each other. I think we already have a pretty good sense that (2) is widely felt, less of a sense on (1), and the final "identify any other and suggest resolutions" clause leaves us wide open to do almost anything.

There are lots of things that could mean for further surveys, starting with more from Amanda's message:

If we think that we want to propose to SIGCSE that there is demand for a standing group, seeing the amount of information that people want about how liberal arts CS education is being structured and that we don't have could be a compelling part of the argument. Additionally, we don't currently have a good place to house any information we gather - hopefully becoming a more long-term group could justify establishing a home for this sort of content.

If we do want to gather more survey data, I might be inclined to assemble more of a meta-survey to build on what we learned at SIGCSE and find out from a larger set of those involved in CS education if we've correctly identified the major issues and concerns faculty at liberal arts institutions have or not. We could likely assemble the survey from the summary document we already have.

To my mind, this says we treat most of the questions raised by the SIGCSE conversation as demonstrating the need for an ongoing body that can conduct studies of liberal arts computing and curate the resulting data over a long period of time. The Committee can then recommend creation of such a body and refer most further surveying to it, maybe focusing any survey and reading we do in the next few months on item (1) from our charge. This might lead us to identify a few new needs from the SIGCSE conversation, but to only offering resolutions to items (1) and (2), both resolutions quite likely being "form a permanent group." I think this is a very reasonable way to go should we choose it.

But another way we could go is to treat the SIGCSE conversation as identifying the other needs that the liberal arts CS community thinks are important, leading us to try to understand and report on all of them. Personally I think this is too open-ended, and takes us way into a gray area of identifying genuine "needs" vs gathering data to answer questions that follow from those needs. The long survey it suggests would be hard to design and hard to get respondents to wade through. But if there was a big turnout of volunteers for this approach, and/or it turned out most of the questions/issues from SIGCSE were already addressed somewhere in existing literature and data, we could try this.

And, of course, there are middle ways between the above: leaving most of what we identified at SIGCSE for a permanent body to pursue, but taking on a few items that we, in whatever way, decide are particularly important now, etc.

So, do others of you have feelings about which of these courses, or others, you'd like to follow?

In all cases I prefer not designing a survey and gathering data that already exists somewhere else though, thus the literature and other data sets search part of the project. (I didn't quote it, but another part of Amanda's message mentioned looking at reports from the NDC and Taulbee surveys and finding that the results don't have a "liberal arts" category per se).

Finally, we do need to produce a report. I think it would be nice for that report to be in essentially final form in time to give an oral version at SIGCSE 2018, with a written one appearing soon after.