For decades, faculty have worked to include social and ethical issues within the computing curriculum. For example, Janet Davis and I presented a paper on "Incorporating Social Issues of Computing in a Small, Liberal Arts College: A Case Study" at SIGCSE 2011[2]. Also, an ITiCSE 2012 Working Group reported on "A framework for enhancing the social good in computing education: a values approach," with specific focus on the inclusion of projects within introductory courses [6]. More generally, faculty experimentation with the integration of technical, social, and ethical issues within the curriculum has continued for many years.

Separately, the Oxford University model of one-on-one tutorials has considerable appeal. A student works directly with a faculty mentor on a project over a semester or longer. Typically, working with a mentor, a student digs into the literature, investigates a topic, and produces a final product, such as a paper or technical report. Although this model helps each student to develop and blossom with expert guidance, each project requires substantial faculty time and thus can seem impractical on a broad scale.

This column presents an experience report, describing a junior-level tutorial/seminar that considers technology and its social/ethical impacts and that engages students following a scaled Oxford-style pedagogy. Based on courses I taught at both Williams College (in Fall 2017) and the University of Puget Sound (in Spring 2020), my experience suggests this approach can work reasonably well with up to 15 to 20 students—not a vast number, but consistent with upper-level courses on many campuses.

As already noted, neither the combination of technology with social/ethical issues nor the Oxford-style approach can be considered new or different. However, in talking with faculty, associate deans, and students on several campuses, this type of tutorial seems quite different from typical courses and thus may seem "new" for a range of departments and curricula.

In these courses, a primary concern is fully engaging students in the consideration of technology and its social/ethical impacts. To illustrate, discussions with fourth-grade students indicate they can recite such principles as "love your neighbor"—the abstract theory seems well established. However, if the same students are asked what happens on the playground at school and how they (or others) behave when going to and from school, it is clear that behaviors do not always follow the principles they recite. Overall, students must engage the interplay of technology and its implications, not simply listen and recite.

Within this context, this column proceeds with six basic sections.

  • Course Goals and Objectives
  • Instructor and Student Decision-making
  • Course Format
  • A Typical Weekly Schedule for a 5-Week Module
  • Additional Notes
  • Course Effectiveness

Appendices A and B include feedback forms for both student posters and papers.

Course Goals and Objectives

In speaking with many computing students, few seem to have much awareness of the consequences of technical applications. The students may be motivated by a desire to help people using technology, but they may have little sense of possible unintended consequences. Thus, I think of this course as immersing students in an unfamiliar environment—generating an interrupt, which forces students out of their comfort level to consider both benefits and potential threats of computing-based applications.

Pragmatically, moderate immersion within a topic seems possible in roughly five weeks, allowing a course to consider about three topics within a 14 to 15 week semester. For each topic, course work examines what a technology seeks to accomplish, how it works, what group(s) it impacts, and what social and ethical consequences might result. With this framework, study of a topic concludes with each student creating a technical paper or poster in the form that computing professionals might present at a disciplinary conference.

This leads to following basic goals and objectives:

  • Goals:
    • Generate intellectual interrupts that challenge students' understandings of the social/ethical dimensions of regarding computing-related applications.
    • Build student awareness of the consequences of technology.
    • Provide analytical frameworks for the consideration of technology and its impact.
  • Objectives (For each topic, students):
    • Develop an annotated bibliography of at least four or five relevant articles, with some addressing technical matters and some discussing social/ethical issues.
    • Present one or more 5-10-minute reports to the class, identifying results of the student research.
    • Review some draft reports from other students, following a review format consistent with a typical conference/journal submission process.
    • Synthesize ideas and conclusions of the work into a paper or poster, following a form and style that could be submitted to a technical conference.
  • Through the course: students should prepare at least one paper and at least one poster, so they will gain experience in both formats used by computing professionals.

Instructor and Student Decision Making

The course goals indicate that students will delve deeply into about three topics, but which topics?

Although some faculty may start a course by focusing on philosophical and ethical principles and approaches, I prefer to jump directly into an application: Big Data—its uses and misuses, using Cathy O'Neil's, Weapons of Math Destruction [6]. This book, a New York Times bestseller often challenges student perspectives, identifies uncomfortable examples, and raises important questions—not as a final word on the subject, but as a fine starting place. Also, the book is widely available and reasonably inexpensive, allowing the course to get started efficiently. For example, this first topic naturally leads to several vital questions:

  • What algorithms and technology underpin neutral networks, probabilistic models, and other techniques within Big Data, and what assumptions typically are made?
  • What formal variables, proxies, and data sets are used in production systems, and to what extent do choices within the development of a model lead to objectivity or bias?
  • To what extent do feedback loops improve a model or reinforce biases or promote the status quo?
  • To what extent is correlation confused with causation, and in what circumstances might individual traits be ignored in favor of socio-economic or ethnic classifications?
  • How might some commercial systems lead to redlining and other undesired/illegal results?
  • What is the impact of current systems on such areas as college admission and financial aid, policing and the justice system, home ownership and insurance, and employment?
  • To what extent can policies, development methodologies, and testing aid or hinder development of fair, useful, and unbiased applications?

In my experience, few computing students have thought about such matters, and O'Neil's book can be quite unsettling. With the course, this book may provide an initial context, but students need additional exploration to consider answers. Faculty may organize course work to address such matters through assigned readings, class discussions, or student-based literature/news searches. Faculty also may elect to introduce alternative moral and/or ethical perspectives. However the course proceeds, students typically find themselves confronting unfamiliar perspectives of technology and its impact.

Beyond this initial topic and resource, the students and I collaborated to choose a few additional topics for exploration for the rest of the course. (See [8] for some details.) In practice, I cannot allow each student to pick separate topics in the pure Oxford model. Rather, in the first offering of the course, I worked with the students to reach consensus on common second and third topics. Although this worked well, students in my second course offering had strong diverging interests, and I allowed two parallel tracks for topics two and three.

Altogether, involving students in the selection of topics (after Big Data) generated considerable motivation and enthusiasm. In addition, although students needed to produce at least one paper and at least one poster during the semester, they could choose which product to complete for each topic—giving them modest control of their own work and activities.

Course Format

This course follows the basic form of an upper-level tutorial, offered at Williams College, in which two students meet regularly with a faculty member for "An in-depth conversation, fueled by intellectual curiosity and the spirit of debate, that takes place over the course of an entire semester," following "the Oxford University style of education." [9] In this format, students are largely responsible for class discussions and activities, with the instructor providing guidance and encouragement—but not taking full control!

My offerings of this course format largely followed the Williams model, with refinements tailored for discussing technological/social/ethical issues. The basic elements follow.

  • Weekly, one-hour, whole-group meeting: Each week, I met the entire group to hear reports from the past week and identify milestones/expectations for the current week. When starting a new topic, I provided a framework and possibly some background. Thereafter, weekly group sessions reviewed the overall schedule, identified forthcoming milestones, fostered student discussions, and provided a setting for student reporting on their research.
  • Weekly, one-hour small-group meetings: From my perspective, the heart of the course is meeting with small groups of students each week. Following the Oxford/Williams approach, students lead these sessions, so the large-group meeting clarifies expectations; and students in small groups present parts of an annotated bibliography, highlight how an algorithm or technical approach works, identify and discuss social/ethical issues, or provide feedback on peers' draft posters and papers.

With this approach, the small groups served as a student/research group. In a typical Williams model, groups consisted of the same two students throughout the semester, although I changed the groups with each topic to broaden a sense of community. Also, to allow larger enrollments and to accommodate differing student interests, at the University of Puget Sound, I allowed groups of three or four, with new groups formed for each topic.


Upon reflection, many considerations for this course reduce to two main themes: student expectations and understanding student backgrounds.


Regarding group size, having four students in a group allows each student to present 10 to 12 minutes each session. Pragmatically, I found I could handle five small groups, so this format allows a class enrollment of about 20. More than four in a group could make it more challenging to include all at a solid level of participation, but such numbers are beyond my direct experience.

A Typical Weekly Schedule for a Five-Week Topic Module

The overall class structure entails investigation of three main application areas, and the course generally follows a similar five-week schedule for each of the three applications. For the most part, each weekly full-group session raises general questions for group discussion and identifies specific expectations for the small-group meetings.

Some modest adjustments may be needed at the start of the semester or when beginning a new application, but the same five-week schedule (repeated for each of the three applications) can work well for each selected topic.

  • Week 1: Orientation and Introduction
    • Full-group Session: Application overview and identification of small-group meeting times.
    • Small-group Sessions: Student presentation of preliminary annotated bibliographies and topic themes
  • Week 2: Technical Issues
    • Full-group Session: Preliminary consideration of some high-level, technical components
    • Small-group Sessions: Review/clarify technical elements (e.g., assumptions, algorithms, processes), what might go wrong, and how might difficulties be addressed
  • Week 3: Social and Ethical Issues
    • Full-group Session: Preliminary consideration of social and ethical considerations, such as stakeholders, impact, and risks.
    • Small-group Sessions: Student-led discussion of specific impacts, risks, implications.
  • Week 4: Start Project (Paper or poster)
    • Full-group Session: Continued discussion of both the technology and its consequences, based on previously circulated questions
    • Small-group Sessions: Students identify a project title (with initial references), and the group brainstorms possible additional directions and ideas that might be considered.
  • Week 5: Finish Project (Paper or poster)
    • Full-group Session: Students distribute to their group a draft poster/paper for their topic and give a 5 to 8 minute presentation.
    • Small-group Sessions: Students bring their feedback on the drafts, with 10 to 15 minutes devoted to constructive suggestions and observations. Discussion might also include reflections on major findings and perspectives.
  • End of Week, Before Next Group Session: Students revise projects (papers/posters) as seems appropriate and submit their final projects.

Additional Notes

Any course requires consideration of numerous philosophical perspectives, practices, and details, and discussion of course elements can extend for pages. (For example, the first draft of this section extended over two pages.) Upon reflection, many considerations for this course reduce to two main themes: student expectations and understanding student backgrounds.

Student Expectations: As this upper-level, tutorial/seminar is different from other computing courses, students need to understand both the intended content and design.

  • Although the course provides a framework for study with the instructor serving as a mentor and guide, students are expected to take the initiative in exploring new areas, identifying resources, leading discussions, and presenting conclusions. This is not a lecture course.
  • While students will study various technical approaches and applications, at least as much emphasis and time are devoted to the impact of the technology, including potential benefits, risks, and unforeseen results.
  • The products resulting from student work will not be computer programs, but rather technical papers and posters, in a form that might be presented at technical conferences.

Understanding Student Backgrounds: An important theme in promoting student success involves tailoring a course to meet the needs of students as they arrive at the start of a semester, rather than building on background that an instructor might hope the student has had. For example, [1] and [10] highlight the diverse backgrounds of incoming students for introductory courses. Since this course differs dramatically from many other computing courses, similar issues can arise here. Two examples follow.

  • Getting started on a topic: Many students have had little experience with research or literature searches, so an instructor might recommend a few initial articles, references, videos or other starting places, or a Science Librarian might provide guidance during a full-group class session. Students may welcome open-ended literature searches after they engaged a topic, but they may feel lost in getting started.
  • Developing a Technical Paper/Poster: Although most students have had experience writing papers for humanities and social studies classes, few have written technical papers or posters, and most need considerable guidance. At first, in-class discussion could outline expected formats and conventions, as well as a process for getting started. (See, for example, [3,4,8].) On-going individual and small-group discussions provided additional guidance and feedback. Some students absorb the styles and approaches quickly, but others struggle and finally understand how to proceed only after detailed feedback on draft submissions.

Course Effectiveness

Although I have taught this course twice, enrollments in both offerings were small, limiting possibilities for a formal statistical analysis of results. Some assessment based on my own experience, however, seems plausible with observations in at least three dimensions.

  • Drop-out Rate: Student enrollment and retention in this course seemed causally related to pre-registration publicity and student expectations. Specifically, in two instances as follow.
    • When students registered for the course with a clear understanding of the content and course format, retention was close to 100%. Students expressed excitement for learning about social/ethical dimensions of computing, and they greatly appreciated the opportunity to gain experience in writing technical papers and posters—anticipating activities in their future professional careers.
    • When students expected the course to rely upon lectures by the instructor with only modest student involvement, it is not surprising that student retention was spotty. Some students appreciated both the course content and pedagogy, and others dropped—indicating they simply did not want to work that hard or wanted to devote their energies to other priorities.
  • Student Intellectual Development: Although students entering the course expressed interest in matters of the impact of technology, student comments at the start of class often reflected an unexamined faith in technology and applications. As the semester progressed, many students next expressed shock and alarm about apparent consequences and misuses of technology. Over time, student comments and writings showed increasing understandings of risks, assumptions, alternative approaches, safeguards, etc. In short, both in-class comments and written materials demonstrated growth in both content and implications.
  • Progression of Technical Writing: Papers and Posters: Although students had written assignments in various disciplines previously, few had written technical materials (posters or papers) previously. Initially, most seemed to feel uncomfortable constructing a technical paper or drafting a poster—they had little notion of how to begin, and initial drafts often were incomplete, awkward, and poorly organized. Again, scaffolding within the course clearly had a substantial impact.
    • Initial discussions within small groups gave students ideas of themes and directions.
    • The submission of preliminary annotated bibliographies and then abstracts helped students focus and refine topics.
    • The submission of drafts brought ideas together—at least to a point.
    • Feedback from peers and the instructor provides multiple perspectives and can have dramatic impact. Appendices A and B illustrate forms that I utilize in this course.
    • Final submissions usually showed substantial improvement over initial drafts, and several submissions were impressive.

Overall, class discussions, submitted technical papers, and developed posters all provided considerable evidence that the course changed student thinking in substantive ways. Simply stated, students talked differently about the impact of technology as the semester progressed. Further, considering draft and final papers/posters as forming a type of portfolio, student writing clearly improved through the semester. In addition, exit interviews with several students indicated that many class members clearly appreciated the assignments in the form of professional submissions and the feedback they received from the instructor and their peers.

As a separate course outcome, posters produced in this course can be publicly displayed (e.g., in a Science Library, an introductory classroom, or a departmental hallway), so ideas from the course can be communicated to prospective students and those in introductory computing courses. In viewing people in these settings, it was apparent that the posters were providing new insights and perspectives beyond the course itself.

Conclusions

Although Oxford-University-style tutorials provide a lovely model for student projects which are guided one-on-one with faculty members, this type of individual student-faculty endeavor does not scale well. For example, with 15 to 20 students involved in a course, individual, hour-long meetings each week, a substantial amount of faculty time would have to be reserved each week. The tutorial/seminar approach described here allows considerable individual attention—but in the context of groups of 3 to 5, rather than one-on-one. This context naturally requires adjustments but has a record of success with classes of 15 to 20 students. Expansion to large class sizes seems uncertain, however.

At Grinnell and numerous other schools, tutorials also are offered for introductory students, highlighting writing, discussion, research techniques, and oral presentation. Although these courses have considerable benefit, the courses themselves have a different feel, at least in my experience. Such a course cannot reach the level of depth possible in upper-level tutorials/seminars, since students have limited background in their understanding of topics, computing background, and general communication skills.

Altogether, upper-level, undergraduate tutorials/seminars can provide a worthwhile opportunity for students to combine technical subjects with social/ethical implications and provide students with experience in creating papers and posters that are part of professional careers in computing.

Appendix A: Feedback Form for Student Posters

I use this general format, sometimes slightly edited, to provide feedback to students, first on their draft posters and then on their final product.

Name:
Comments on the final Poster on [topic/title]
Grade:
Overall Comments
Content:
        Identification of key points:
        Statement of main theme(s):
        Support of theme(s):
        Clarity/effectiveness of presentation:
Display:
        Logical organization/flow:
        Balance of components:
        Visual interest:
        Readability/clarity of expression:
        Acknowledgments/Conclusions:

Appendix B: Feedback Form for Student Papers

This form, sometimes slightly edited, provides feedback on draft posters and the final product.

Name:
Comments on the final Paper on [topic/title]
Grade:
Overall Comments
Abstract
Overall Structure/Logical Sequencing of Topics
Introduction
Development
Acknowledgments
Conclusion
References
Notes on Writing

Acknowledgments

This column emerged from my experiences teaching an upper-level, seminar/tutorial-style course at Williams College and at the University of Puget Sound. Many thanks to those involved with the academic program at Williams for establishing this type of course as a regular course offering and providing considerable guidance. Special thanks to Brent Heeringa, Duane Bailey, and Andrea Danyluk at Williams and to Brad Richards and David Chiu at the University of Puget Sound for their guidance, support, and insights as these courses were being proposed and developed. Without their help, these courses could not have been offered! Thanks also to the students who participated in these courses and provided numerous comments and suggestions. Additional thanks to the reviewers of this column for their feedback regarding the organization and level of presentation for the various themes presented in this column..

References

1. ACM Education Board Retention Committee, Retention in Computer Science Undergraduate Programs in the U.S. Data Challenges and Promising Interventions, 2018; https://www.acm.org/binaries/content/assets/education/retention-in-cs-undergrad-programs-in-the-us.pdf. Accessed 2020 July 9.

2. Davis, Janet, and Henry M. Walker, Incorporating Social Issues of Computing in a Small, Liberal Arts College: A Case Study, SIGCSE '11: Proceedings of the 42nd ACM technical symposium on Computer science education March 2011, 69–74; https://doi.org/10.1145/1953163.1953186. Accessed 2020 April 26.

3. Ernst, Michael, How to write a technical paper; https://homes.cs.washington.edu/~mernst/advice/write-technical-paper.html (updated October 20, 2019). Accessed 2019 October 23.

4. Ernst, Michael, Making a technical poster; https://homes.cs.washington.edu/~mernst/advice/poster.html (updated October 18, 2016). Accessed 2019 October 21.

5. Felten, Edward W., et al, Lecture 1: Intro. to Crypt and Cryptocurrencies, the first of a multi-part lecture series on Bitcoin and Cryptocurrencies; https://www.youtube.com/watch?v=fOMVZXLjKYo. Accessed 2020 May 23.

6. Goldweber, Michael, John Barr, Tony Clear, Renzo Davoli, Samuel Mann, Elizabeth Patitsas, and Scott Portnoff, A framework for enhancing the social good in computing education: a values approach, ITiCSE-WGR '12: Proceedings of the final reports on Innovation and technology in computer science education 2012 working groups, July 2012, 16–38. DOI: https://doi.org/10.1145/2426636.2426639.

7. O'Neil, Cathy, Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy, (Broadway Books and Crown Publishing, New York, 2017). Shoop, Libby, A Guide for Writing a Technical Research Paper; https://www.macalester.edu/~bressoud/capstone/TechPaperHowTo.pdf. Accessed 2020 May 23.

8. Walker, Henry M., Algorithms and Applications: Opportunities and Risks, CSCI 395 at the University of Puget Sound, Spring 2020; http://www.cs.grinnell.edu/~walker/courses/395.sp20-ups/. Accessed 2020 April 26.

9. Walker, Henry M., with the ACM Retention Committee, Retention of Students in Introductory Computing Courses: Curricular Issues and Approaches, ACM Inroads, 8, 4 (2017), 14–16.

10. Williams College, Tutorials (Home page for Williams College tutorials), https://www.williams.edu/academics/tutorials/. Accessed 2020 April 26.

Author

Henry M. Walker
Department of Computer Science
Grinnell College
Grinnell, Iowa 50112 USA
[email protected]

Figures

F1Figure 1. Many factors come together in the described tutorial/seminar

Copyright held by author/owner.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2020 ACM, Inc.

Contents available in PDF
View Full Citation and Bibliometrics in the ACM DL.

Comments

There are no comments at this time.

 

To comment you must create or log in with your ACM account.