Teaching Responsible Computing Playbook

Making Lessons Stick

Authors: Helena Mentis, Vance Ricks

As we saw in Choosing Computing Courses section, courses are the primary mode of interaction between faculty and the vast majority of the undergraduate students and so much of the guidance provided in these chapters are about single courses. But when thinking about how to ensure knowledge and perspectives persist beyond the classroom, it helps to think about how classes build on one another and reiterate the responsible computing knowledge and application across the curriculum. As a goal, witnessing the progression of thought is better than expecting a student, especially new undergraduate students, to understand and be able to address every ethical dilemma in just one semester. The intended audience for this section is a department (or a subset e.g. group of faculty who are interested in responsible computing) rather than an individual faculty.

The “[X] across the curriculum” models challenge the academic division of labor by which a particular discipline/department is supposed to teach “skills” and a different discipline/department is supposed to teach “content”. For example, a Writing Across the Curriculum model challenges the idea that the Composition, English, or Writing program is where students learn the writing “skills” that they will then apply to the “content” they’re learning in, say, their coursework in the sciences. That idea presumes that “good writing” consists of purely formal or technical elements that are context- or task-independent. Instead, a Writing Across the Curriculum model begins from a more substantive, contextual, and task-oriented understanding of “good writing”. It doesn’t deny that there are formal elements common to all good writing. However, it emphasizes that students should learn their major discipline’s standards of good communication (including good writing) from suitably-trained instructors in that discipline, perhaps in specially designated “writing-in-the-major” courses. This means that students are learning that learning to write/communicate well is a priority, both in their general curriculum coursework and in their major-specific coursework. Applying that model here, to responsible computing concepts and habits, means not subcontracting the teaching of all of those concepts or habits to some other department, but instead thinking about a relatively small number of core things [methods, skills, concepts, habits] that the students will encounter repeatedly throughout their coursework in computing. As in the case of writing-in-the-major courses, the responsible computing-focused courses should include concepts that are authentic to the discipline or area.

In thinking across the curriculum, the computing education team must scaffold learning outcomes just as they do with programming. For instance, in programming classes, students must learn there are integers before they learn how to assign a value or even how to call a database to assign that value. This is the same with ethics and responsible computing knowledge - students need to first learn that there are different people with different and evolving lives and experiences before being able to realize that hard coding gender variables may present a problem to inclusion or to ever understand that using a training set of Google images to train a gender identification algorithm may be a problem.

An analogy to hold as the team is thinking about this scaffolding process is to first plant the seeds in the first year and then continue to water them throughout the curriculum until graduation. There are two key aspects to this model. First, it is important to plant the seeds early in the curriculum. The idea that one can just drop ethics and socially responsible thought on a third or fourth year computing student and suddenly they will see the connections between technical and social decision making is asking a lot of anybody, let alone undergraduate computing students. It is incumbent on us to seed the ideas early - preferably in the students’ first semester. The second key aspect of this model is that students are not going to be ethical experts after that first semester. The goal of that first semester is to turn on the switches in their head and then to reiterate and build on the ideas semester after semester. Each subsequent class a student takes then activates the prior knowledge and adds more depth to it. In addition, each semester’s technical content can be further connected to the relevant ethical perspectives.

The goal with these two key aspects is that, after four or so years, a student not only will be technically proficient but also understand the social responsibility and impact of technology as being integrated into every decision one can take in their design and development process. The students have had multiple opportunities to apply ethical thinking to technical design and development as well as multiple opportunities to learn how to advocate for ethically sound goals. A seeding model raises the likelihood that this integrative thinking is second nature for the new computing graduate as opposed to some secondary thought process that requires additional cognitive resources. By making ethical thinking a fundamental way of grappling with technical decision-making and work throughout one’s four-year education, it is far easier to enact ethical and responsible computing thinking processes when in the workplace. And it will be much easier to see the actionable options before them when faced with ethical challenges in the workplace as one of the best ways to argue against unethical practices is to provide ethical solutions as suitable alternatives. That is a skill that needs to be practiced and reiterated just as much as proper coding practices.

A related aspect, the way the responsible computing assignments activities are structured is also very important. If the students superficially get exposed to responsible computing material without having to go through a detailed thought process when thinking about responsible computing, then even getting exposed to such exercises in multiple courses will not be as beneficial. Among others, such assignments should challenge students to revisit any held beliefs their life experience is universal, and encourage them to perceive themselves as the “other.”

Indeed, the usefulness of this model is that it relieves the pressure off of one or even a small handful of classes to be the watering cans in integrating the ethics curriculum. The computing curriculum is already packed and there is already such a breakneck speed that it is not allowing the time and space to be able to grapple with these bigger issues. By any given class only needing to grapple with one or two concepts deeply, it increases the likelihood of more faculty buy-in as well as helping the lessons stick. In addition, there are watering cans that exist outside of the classroom that can be seen as a part of the full model: peers (through a cohort model), student groups (e.g. CWIT at UMBC, NSBE), student activities (e.g. socially responsible hackathons with domain expertise), faculty interactions, research experiences, internships, etc.

Key Questions:

What are the best first semester classes to choose to start to plant the seeds? Multiple first semester classes can be used to reach all majors and experience levels - as long as every student can be exposed to the seeds in their first semester.
What classes are the best choices for each subsequent semester? Looking at a typical student’s progression through their intended major, at least one class each semester could be identified as the watering cans.
What are the relevant student learning outcomes for the targeted courses? See the section on Learning Outcomes and Assessments for more on this.
Do the in-class activities/assignments make students challenge their own beliefs and/or make them go through some difficult introspection? Having students work in diverse groups to come up with a common response to a real-life situation could potentially help with these. See the Difficult conversations section for more on this.
Do the graduate and/or undergraduate teaching assistants/fellows have the prerequisite seeds to provide support? Depending on the order in which students take classes, undergraduate teaching assistants may not have the proper seeding yet and graduate teaching assistants may not have been seeded in their prior programs.
What other activities at the university can be the watering cans? Encouraging student groups to reiterate ethics concepts as well as facilitating coop and internship placements at industries that support ethics and social responsibility thinking.

Checklist

☐ Identify all first semester classes that are suitable seed classes.

☐ Identify at least one class per semester that can be used as a watering can (based on a typical student’s progression through the curriculum).

☐ Map the ethics/responsible computing curriculum outcomes to the classes identified above.

☐ Design in-class activities/assignments that challenge students’ beliefs and have them work through uncomfortable scenarios.

☐ Support faculty and teaching support staff in identifying which seeds can be watered in their particular class.

☐ Identify other activities that can be used as watering cans.

☐ Continue to coordinate with these opportunities for reiteration.

University of Maryland, Baltimore County

At the University of Maryland, Baltimore County (UMBC), we did the seeding in COMP101 and continued with related CWIT activities.

Checklist walkthrough. We walk through the following checklist items that we worked out at UMBC:

Identify all first semester classes that are suitable seed classes.
- COMP 101 is the first programming class taken by many computer science, information systems, and business technology administration students. We seeded the class with discussions of, for instance, representation and diversity in computing practice and technology use. This turned out to be one of the most successful lessons that ‘stuck’ as evidenced by our end of semester focus groups and 6 month later follow-up interviews.
Identify other activities that can be used as watering cans.
- In addition to other higher-level courses that were integrating aspects of ethics and socially responsible thinking (e.g. the integration of ethical concepts into undergraduate Data Science classes at UMBC), there are other activities and groups in and around the undergraduate curriculum that can reiterate concepts and values. For instance student groups often host speakers and coordinate activities like hackathons that can highlight ethical considerations. At UMBC, the Center for Women in Technology centralizes the importance of inclusion in computing - both in who can participate in computing as well as how to create inclusive computing technology. Thus, in a student’s first year, the lessons were not solely in the COMP 101 class. These topics were also being talked about outside of the classroom through CWIT events and cohort groups. This is an example of creating the ecosystem where we are reiterating lessons - just as we were reiterating the importance of what CWIT was saying, CWIT was reiterating the lessons we were integrating with, for instance, their Living Learning Communities. This is because it is deeply tied to their Values.
Continue to coordinate with these opportunities for reiteration.
- To help students make these connections, it helps to have a coordinating body to see the connections between different efforts and ensure coordination. For instance, ensuring class instructors are aware of invited talks and student activities that can be highlighted and reiterated in the day’s lesson. In addition, a coordinating body can ensure that there is a relationship that naturally progresses between topics taught at the early introductory classes to those in subsequent classes. At UMBC, as of 2020, the Dean’s Office is facilitating this coordination, but through a new internal education innovation grant, a team has been formed to facilitate this coordination across the college as well as with the humanities and social science scholars that also teach the university’s computing students.

University at Buffalo

At University at Buffalo (UB), we targeted multiple courses throughout the UG curriculum to include responsible computing content. We also designed activities that “forced” students to think through some activities with no “obvious” answer.

Checklist walkthrough. We did the following at UB:

Identify all first semester classes that are suitable seed classes.
- We targeted CSE 199 our first-year seminar. Module on history, society and diversity and module on societal computing were targeted. Each module is a two week-long combination of lectures and in-class activities.
Identify at least one class per semester that can be used as a watering can (based on a typical student’s progression through the curriculum).
- The following courses were targeted (corresponding to courses typically taken at the sophomore, junior and senior years of our BS Computer Science major):
  - Algorithms course (CSE 331): Homework assignments and a project based on access to high-speed Internet in WNY were introduced. CSE 331 is required for the BS Computer Science program.
  - Machine learning course (CSE 474): Students worked in a group project exploring bias in data-driven criminal risk assessment.
  - Capstone course (CSE 442): Students review a made-up dossier of government data to identify if an individual should be added to the government’s No-Fly list. Students must then discuss biases in the data provided and, finally, how they would respond if their job assigned them to the team automating this analysis. A second assignment asks students to document the ethical implications of the capstone project they chose. CSE 442 is the capstone course for the BS Computer Science program.
Map the ethics/responsible computing curriculum outcomes to the classes identified above.
- We created a new grading rubric for ethics-related assignments, which were also useful in identifying the student learning outcomes.
Design in-class activities/assignments that challenge students’ beliefs and have them work through uncomfortable scenarios.
- Through trial-and-error, we found that the questions/activities/experiences that work best are ones which diverse groups work to develop a single answer (this helps avoid the issue of traditional students assuming that their life experience is universal) and force the students to make decisions and then evaluate how those choices reflect their ethical choices (this helps with the issue of traditional student being unable to perceive themselves as the “other”). E.g. this is done in the no-fly-list exercise in CSE 442.
Support faculty and teaching support staff in identifying which seeds can be watered in their particular class.
- The Mozilla RCS award funded the faculty time but even more importantly funded a dedicated undergraduate TA pool for each of the four targeted courses in the proposal (first-year seminar, algorithms, machine learning, and the senior capstone course).