My Proposal for "It Seemed Like a Good Idea at the Time"

Next week, I will be presenting in the "It seemed like a good idea at the time" panel at SIGCSE. Each of the panelists submitted a 150-word summary of their experience that will be published in the proceedings. This is mine:

When the pandemic left a colleague unable to complete teaching an upper-division programming languages class, both sections were given to me for the last two weeks of the semester along with the mandate to give the students "a good experience." The previous instructor left no plans for the end of the semester, and I had not taught programming languages since the early 2000s. I saw the opportunity to give a brief introduction to declarative programming through Prolog. I asked the students to complete the classic family tree assignment. The unfortunate result was that over 25% of the students violated the university's academic integrity policy, submitting solutions copied from each other and from the Internet. Three days of paperwork was not how I expected to wrap up a stressful academic year, but the experience exposed interesting conflicts in culture and expectations between me and my students.

It's fine. 

The reason I got to write it—the reason I will be able be part of the panel—is that I wrote a proposal to do so. As I prepare my presentation for SIGCSE, I have returned to the proposal, which was written shortly after the experiences described in the abstract. As I feared, it is a bit like opening old wounds, since it was a stressful and spiritually draining experience. That said, I had spent significant time working on the proposal, even though it is not itself published. 

After some consideration, I have decided to re-post the proposal here with just one minor correction. The section headings were given as part of the required format. The task for me now is to determine how to take the ideas contained within the proposal and distill them into an eight-minute presentation.

A description of the good idea, including your inspiration, how it was implemented, and the expected results

During the Spring semester, the impact of the pandemic left one of my colleagues unable to successfully complete teaching his classes. Hence, two weeks before the end of the semester, I was assigned to take over both sections of his programming languages course. It is an upper-division required course here, and this was the first time it had been taught online and asynchronous. There were no plans for the end of the course, nor the final exam; I had carte blanche and a mandate from the department chair to give the students a good conclusion to the course.

I never had taught the course here, and the last time I taught it was in graduate school in the early 2000s. Reviewing the department's syllabus and the previous assignments, I saw that the students had not yet studied declarative programming, which is one of my favorite paradigms. I decided to give them a quick, one-week introduction to Prolog, just showing the simplest examples of facts and rules. For the week's assignment, I gave them the definition for several family relationship rules (father, sibling, etc.) and asked them to write several more. I explained in the assignment that this was a "classic problem" in Prolog, and that they could doubtless find answers to these challenges online, but that they were being asked to puzzle through the logic and sort it out themselves.

The expected result would be that students would be intrigued by this unconventional programming paradigm, puzzle over the challenge, email me a few questions about it, and enjoy it as an end-of-semester curiosity. It was not designed to be very hard, and so I expected that the high grades on this straightforward assignment would raise their course grades, which is something students always seem to like and that seemed appropriate, given their tumultuous experience during the rest of the semester.

A description of the actual results

A shocking number of students violated the university's academic integrity policies on this assignment. I followed up on over 25% of the students for violations. The number I discovered with relative ease is a lower bound on the possible violations, modulo a very small number of false positives. Most of these students copied solutions verbatim, either from the Internet or from each other; some of these even copied and submitted solutions that did not meet the assignment specifications.

My morale was devastated by this experience, which came at the end of a stressful academic year, and I might have thrown in the towel if it weren't for one case. Most students were silent about the academic integrity violation, and a few students wailed and wheedled over email and Zoom, but one student stood out as a pillar of integrity. This student sent me a message admitting their guilt and, essentially, thanking me for catching them. They explained that they knew what they did was wrong, and they were sorry for it and accepted the results with a clear conscience. This student gave me hope that, indeed, there are still those who understand the educational value of these policies and that the lasting goal is not certification but the acquisition and practice of virtue.

A simple description of the evidence supporting failure that you will be able to supply

The percentage of academic integrity violations is the evidence of failure. I have spent a lot of time pondering how it came to be, and my best hypothesis is that the real failure was one of mismatched expectations and culture. I share some of these in the section below about the presentation. If only I had known this would happen and could have collected qualitative data about student perceptions—what a research study this would have been!

Ideas for your presentation

The presentation would have to begin with the simple narrative of what happened. Once they understand the story, the audience should be able to follow my reflections on the experience. I want to discuss the various mismatches of culture and expectations that are represented in my story. Briefly, the two major points I want to make are (1) differences in expectations and (2) the transactional vs. interactional nature of higher education.

One mismatch comes from my experience learning Prolog and theirs. When I studied and previously taught programming languages, there was a young World-Wide Web, but we could not have had the Copy-Paste Programming attitude that rose to prevalence with sites like StackOverflow. The entire K-12 experience was different for me than it is for contemporary undergraduates, who have been inundated with dehumanizing standardized tests. I suspect that current students are not only more interested in "the answer" than past students, but that they also see this itself as virtuous: that is, I think many believe the answer is the goal rather than the pursuit. Put another way, it was not at all clear to me that most of these students understood what they did as "wrong," and they considered the real problem only to be that they got caught. This attitude is reinforced by schooling, including higher education's constant promotion of job-readiness and certification culture. At the risk of offending zealots, I believe that the university's idea that any course can be put online sends the message that there's nothing valuable about what's happening among humans in the classroom. Yes, some people can teach online well, and many people teach in-person poorly. My point is that students are inundated with a subtext that says "seek ye first to get your degree and get a job and make money."

This would be no different here than in any of my other classes except that we have to look at another mismatch, to which I alluded above: that the students had no personal relationship with me nor with the previous instructor. For most of the semester, the previous instructor struggled to stay on top of the class. When I took over, almost none of the students had ever met me. Despite my attempt to welcome and encourage them through written messages and talking-head videos, there was no real opportunity for us to build a culture of trust. This is in contrast to my usual courses, where I pride myself in having high standards and in building a classroom culture in which students recognize that they can meet them. Instead, this programming language class was instituted as transactional rather than interactional, and its modality—along with so many other students courses—emphasized this as well.

In many ways, this is a sad story with no real conclusions, but as a scholar, I see many opportunities for further research. I would like to end my presentation by encouraging the audience to consider designing qualitative studies to better understand these cultural issues, and for us, as a community, to think deeply about what it means to teach students to live "the examined life."

(The proposal concluded with the 150-word summary that I have already shared above.)