This essay suggests teaching engineering ethics by having them write case studies based on their own projects.
Engineering, like medicine and law, is practiced within a broad societal context. As such, it tacitly incorporates shared societal values, yet also reflects particular demands derived from the practice itself. These tacit values are not easily identified by practitioners, nor are the ethical dilemmas that arise from conflicts between them. The course discussed here is designed to help bioengineering students, both graduate and undergraduate, identify the values embedded in their practice. It stresses the acquisition of methods of moral problem solving to be used to identify, analyze and resolve these dilemmas as they confront them.
The course assignments and the methods of student evaluation have been enriched by "lessons learned" from an NSF grant one of us has been involved in for the past four years1. Below, we describe and evaluate one such aspect of the course: having students write and analyze their own case study. In our grant, we characterized ethics as an "ill-defined domain" and this led to several pedagogical approaches to teaching. One such approach was to have graduate students write their cases based on their research and undergraduates to base cases on their senior design project. Reasons for this approach and an evaluation of it are described below.
Back to Top
Using a popular engineering text, Engineering Ethics: Cases and Concepts by Harris, Pritchard, and Rabins, select readings from medical ethics and another engineering text, Engineering Ethics: Balancing Cost Risk and Schedule by Pinkus, Shuman, Hummond, and Wolfe2, the course is taught using a case-based approach3. It introduces students both to paradigm cases and to common cases as reference points for learning ethical reasoning. The goals the course are to:
While all students have personal methods to resolve moral problems, the course stresses the importance of learning professional ethical principles and other guides. Conceptual tools such as line drawing, methods for resolving conflict problems, principles of common morality, creative middle way solutions, considering rights of persons and three organizing principles of competence, responsibility, and a revised version of Cicero's Creed (Pinkus, et al. 1997). are presented as ways to help students identify moral issues and proceed through the steps, of analysis. The resolutions students come up with are not "ideal" resolutions, in a philosophical sense, but ones that have been reflected upon and then negotiated in the practical arena.
Students are encouraged to gain knowledge of "ethics in practice" by observing at a hospital ethics committee meeting, attending an Ethics Grand Rounds and taking a "guided" field trip to Washington, DC to observe an FDA panel meeting. The class is required for both graduate and undergraduate students and is taught over a 14 week time frame by an experienced ethics teacher and several visiting professors. Two sessions are taught by co-director of a division of the FDA who hosts the graduate students when they observe the FDA panel meeting in D.C.
Evaluation in the course is based on student's class participation (10%) and two short reports on their out of class observations. (15%) Students have a number of related assignments for the class. They write their own case study, which they present to the class for peer review and discussion (25%). After the presentation, they write a paper in which they analyze the case using methods learned throughout the course (50%). All students are encouraged to use their "moral imagination" in constructing the case (i.e., it is to be centered on their technology of choice but can be embellished to depict a personal, professional, or societal ethical dilemma).
The decision to have students focus on their research or senior design project to write their case study was informed by results of the NSF grant cited earlier5. In that grant, we acknowledged that the process students use in learning from the case-based texts "cannot be reduced to a mechanical procedure or algorithm." As Vivian Weil explains, "Reflection and deliberation about what to do are based on comparison of cases and agreed upon moral standards, as well as engineering codes of ethics. In reasoning about cases, it is advisable to identify central clear cases on which there is general agreement, but not to neglect less clear cases, once it has been shown that there are sometimes right answers... Familiarity with a range of cases is valuable for learning to reason by analogy across cases, identifying clear central instances and distinguishing so-called gray area examples." [Weil, 1992, pp. 5,7].
Practical ethical reasoning in engineering, in sum, involves solving problems in complex, realistic scenarios without the benefit of algorithmic or mechanical formulations of problem-solving processes. We recognized that making decisions in or solving engineering ethics problems is related to two bodies of literature in cognitive psychology: solving ill defined problems, and learning principles of a domain from studying examples or cases. Cognitive psychologists differentiate between well-defined versus ill-defined problems. Well-defined problems are the kind we typically encounter in the classrooms, such as geometry or physics problems. They are characterized by having explicit correct answers that can be justified by the application of a specific principle. Ethical problems, being ill-defined, differ from well-defined problems in three key ways. The first difference is that ethics problems do not have a definitive answer. The answer depends in part on the justifications, as well as the second difference, namely, that it depends on how the problem is conceived. This latter difference means that a given problem, as posed, can become a different problem depending on what constraints and conditions a solver adds to the problem.
Put another way, an ill-defined problem usually requires that the problem solver define it better by adding constraints to the problem, perhaps also adding permissible operators (such as coming up with alternative actions). For example, in order to apply the principle that "Engineers shall perform services only in the areas of their competence", where competence is defined as "qualified by education or experience in the specific technical fields involved", a great deal of interpretation is required in deciding what competence is. Is a 2-year degree "qualified education"? This condition depends on circumstances. Presumably, in an emergency situation, the decision maker may be the best qualified person under the circumstances. A student may supply this supposition in constraining the problem to interpret the condition. By adding such constraints and conditions, the student in effect is solving a different problem than the one posed.
Solving a physics and an ethics problem also differ in a third way. In order to resolve many ethics problems, the issue is not only retrieving appropriate principles/codes, but also mapping the concept in the principles/codes to the situation at hand, at least in cases that are not clear-cut. This mapping process requires a lot of subjective interpretation. In such cases, where a subjective interpretation is needed, a way to make such decisions might be the retrieval of analogical cases illustrating what was deemed "material" in other circumstances. Thus, to solve ill-defined problems, students have to learn to retrieve and compare with analogical cases and to specify constraints on conditions. As a result, there is a much more serious memory component involved in solving ethics problems, (than well-defined problems) in both the retrieval of related cases and the retrieval and operationalization of the principles.
Defining ethics as an ill-defined domain and understanding the general ramifications of this definition led to the following pedagogical technique. It seemed reasonable to assume that one way to have students learn the methods of practical ethics was to have them apply it to a domain that they were familiar with, thus their senior design project or research focus. Note that during the first and second years of the course (and prior to the NSF Grant work) students could write their case study/paper on any topic that interested them. During the third year, (post-NSF) they had the choice of using their research topic or not and in the fourth year, students were required to do so.
Below are the results of an analysis of the 48 cases written and analyzed by 12 undergraduate and 38 graduate students. The papers were written during the four year time frame that the class has been taught. They were evaluated to assess how each student learned the methods of moral reasoning taught in the course. While five methods were stressed in class6, we coded for only one of these: students' use of individual concepts. Our definition of "concept" incorporated broad theories such as utilitarianism as well as specific concepts such as bribery . These concepts were compiled from the both the bold and italicized terms in the Harris text. and those used in the Pinkus text as well. There was a total of 42 concepts. (see Appendix A for full description).
The concepts were coded in three ways. The first focused on whether the student used each individual term. The second scored for whether the student correctly defined or in some way demonstrated understanding of the concept. The third was a combination of these two. In this way we quantified the student's understanding of the ethical concept. For example, if a student listed the term autonomy, risk/benefit analysis, doctor-patient relationship, and informed consent, and then correctly defined or demonstrated understanding of each term except informed consent, the scoring chart would resemble that shown in Table 3. Using this method, the highest score a student can achieve is 1. Table 3 demonstrates the method we used.
Coding in this way evidenced a clear trend towards increased concept learning when a familiar domain is used. (Note: at this point in our data collection we are not claiming statistical significance, merely a trend.) Students who voluntarily chose to use their own research scored higher than any other group. Presumably, they had already attained a working knowledge of the area. Requiring students to use their research points them in a direction they should be knowledgeable about, but of course does not assure "expert knowledge". Nonetheless, Students who were required to choose their research as a topic for the case did better than those who worked on topics unrelated to their area of specialization.
Graphs 1 through 4 illustrate the "top ten" concepts that were used by all students and then show their frequency of use by each individual group: undergraduates, graduates and the graduate students who chose to use their research to write their case, those that did not and those that were required to. Note: there are four separate concepts that deal with safety or risk assessment: safety, risk, cost-benefit and risk - benefit. Each is defined slightly differently in the texts so we separated them in our analysis. But it is important that combined, these account for over 1/4 of the focus for graduate student cases (37% for those who chose to use their research for the case study) and 17% for undergraduate students.
It is important to note that the goal of the course was to teach Professional Ethics' to help the students attain a "toolkit" of methods to apply when they are faced with moral a dilemma that occurs in practice. This goal is akin to what Caroline Whitbeck has termed "agent- centered" learning; or preparing "students to address moral problems that are likely to arise in their work as engineers and scientists' (1995, p. 300). This was not, in short, a survey course to introduce students to special topics in bioengineering ethics or an analytical philosophy course to teach them ethical theory. We conceptualized ethics in cognitive science terms and described it as an "ill-defined domain". As such, we concluded that when teaching a case-based method of moral reasoning that also includes some principles and guidelines, students would be required to master a serious memory component in both the retrieval of related cases and the operationalization of appropriate principles. Having students write and analyze a case that focused on a technical area in which they were proficient allowed them the time necessary to concentrate on learning the "ill-defined domain" of ethics.
Undergraduates benefited from focusing on their senior design project as this enabled them to access a concrete bioengineering topic that they were familiar with and also compensated for their general lack of practical work experience. For graduate students, having them focus on their research reined in both their focus and their enthusiasm a bit, as many expressed an interest in writing a paper examining the ethics of "headline" ethics topics such as cloning or the genetic modification of food. Once they began to examine their own research, however, they were able to take an in depth look at their work from a different perspective. As the accompanying examples show, the results of such an in depth look can be quite impressive (see Appendix B). Students may find topics other than their research focus to be more "engaging" initially, but the assignment thus far has received positive comments in the formal evaluation of the course.
Teachers of practical and professional ethics have long recognized that students with practical experience in their fields tend to learn methods of moral problem solving "better" than those who are novices. In teaching ethics to medical students, for example, Clouser and others advocate scheduling classes in the third or fourth years when students are involved in their clinical rotations7. In these seminars, a case-based rather than a theoretical, principle-based approach is favored during the first half of the course. The idea is to get students involved in discussing issues, grappling with the facts and concepts embedded in particular cases, and then, when they are "hooked" on the challenge of resolving the dilemmas, introduce them to more abstract methods.
Why is it that experienced practitioners tend to learn ethical problem-solving more easily than novices in their field? What lies behind the lure of cases in teaching applied ethics? This paper described a course assignment that used insights from cognitive psychologists study of "ill-defined problems". It also reported the results of an analysis of 48 case studies and analyses written by graduate and undergraduate students. We conclude that if one's goal in teaching professional ethics is to equip students with a method to identify and analyze moral problems in a specific technical domain, having them write and analyze their own cases based on a knowledge domain with which they are familiar, can facilitate their learning.