The Ethics of Evaluating Instructional Computing - Optimum Instruction - The Research Center on Computing & Society

Text-only Version

Marvin J. Croy

3.2 Optimum Instruction

Another ground for rejecting the appropriateness of subjecting students to different treatment conditions in the classroom concerns the responsibility for providing optimum instruction. Overall affirms this responsibility while Moor speaks merely of providing “adequate” instruction. Nevertheless, it is not clear that accepting this responsibility precludes controlled studies of the sort considered here. Conversely, it may be that a commitment to optimum instruction requires rather than prohibits controlled evaluation studies in the classroom. This possibility is suggested by the view that what does or does not constitute optimum instruction is empirically determined. That is, a commitment to optimum instruction, unless hollow, is a commitment to gathering empirical evidence in the classroom. The question is whether that evidence must come in the form of controlled experiments. Overall’s resolution maintains that there exists an adequate alternative.

The value of the CAI programs could perhaps more easily and certainly more fairly have been assessed by providing them for both sections of the logic course, and then comparing their outcomes with those of students enrolled in the logic course in past years. There is no reason to suppose that this year’s crop of students is appreciably different from those of previous years so the instructor has adequate evidence to permit her to evaluate the differences in student achievement produced through the use of CAI programs.(14)

Overall believes that this sequential approach is more fair than an approach which assigns students to a control group. Her resolution, however, is not completely without problems. On the one hand, it is true that current students have not been deprived of what Overall terms “an apparent educational benefit.” But given the genuine possibility that the CAI treatment may be of no value or even counterproductive, this advantage is of limited significance. In fact, the current students have been subjected to the same risk endured by experimental subjects in a controlled study, and Moor’s resolution may be required should current students underperform previous students.

There is also some methodological difficulty with Overall’s resolution. Students can in fact differ from one semester to the next in respect to their ability to learn a particular subject. Consequently, a simple comparison of one semester’s students with those from past years, can easily lead to a faulty conclusion concerning the efficacy of a pedagogical innovation. A dedicated instructor, however, might track student performance for several semesters and then do the same for several semesters following some innovation.(15) In addition, it would be helpful to develop a pretest or to establish a high correlation between some measure such as SAT scores and the ability to learn the subject being taught. These measures could document the similarity of the groups compared. While this would be an improvement, this approach, according to accepted cannons of research methodology, would still be more liable to error than an experimental control group study. This raises the question of how reliable the evidence must be when evaluating instructional techniques. In general, the need for reliable evidence increases as does the extent to which well established practices are modified. An innovation which radically changes the instructional process requires thorough evaluation. CAI programs vary in this respect. Some programs function merely as homework exercises and provide automatic grading while others alter the patterns and degrees of student-teacher interaction. Judgments concerning the reliability of the required evidence should be made by particular instructors considering particular programs and innovations. In any event, a commitment to optimum instruction does not rule out controlled evaluations nor does it establish the acceptability of other alternatives.(16) To provide optimum instruction is to do the best we know how to for students. Characterized in this way, the epistemological burden is obvious particularly since our understanding of effective instruction is continuously evolving. Determining what constitutes optimum instruction is best accomplished by empirical studies over long periods of time. The general nature of the research required bears on certain claims elaborated by Overall. Specifically, the student view that participating in controlled studies “will benefit not them but only some hypothetical future students” is worth comment here.

Educational improvements are best achieved by long term, sustained processes of change involving feedback mechanisms for the evaluation of innovative methods. These improvements are the products of long run commitments to what is basically a process of guided trial and error. One consequence is that, over time, some students are treated differently than others and some students receive better instruction than others. While this consequence does not in itself legitimize the different treatments required by controlled studies, it is germane to the dim view of “hypothetical, future students” taken in the students’ position. The quality of education enjoyed by current students has grown out of years of trial and error. Students cannot expect quality education without acknowledging the contributions of previous students (even though few of those students served as subjects in controlled experiments). Nor can today’s students justifiably demand optimum instruction without accepting their role in that process of trial and error. Students are not, by mere virtue of this role, obligated to serve as subjects in controlled experiments. But by the same token, the fact that only hypothetical future students may benefit from the experiment carries little or no weight in the assessment of that obligation.

Another component of the students’ position worth comment is that which is based upon their role as consumers of education. No student wants to be deprived of products and services available to others when all students pay the same fees. Nevertheless, the view of students as consumers of education is misleading when determining their rights in the context of controlled studies in the classroom. It is true that when selecting an institution of higher education, students, usually in conjunction with their parents, often think and behave as consumers. They think in terms of what will be received for what is paid. They make comparisons of campus living conditions, the surrounding environment, characteristics of students and faculty, ratings of various programs, etc. Nevertheless, they are not acting in a consumer mode when they “pay” for their education. An education is not something which can be bought. It is not a product which can be purchased (like a house or an automobile). Nor is it straightforwardly a service (like having one’s yard mowed or one’s car washed). Normally, if one pays for a product or service and gets nothing in return, one has good grounds for complaint. But it is commonplace in higher education for a student to pay tuition, learn nothing, receive a grade of ’F’ with no academic credit, and yet have no grounds for complaint. In paying tuition the student has not purchased credit hours or a grade in a course. What the student has paid for is an opportunity to be instructed, an opportunity that brings with it numerous responsibilities. To receive academic credit the student must perform assigned tasks, fulfill course requirements, and ultimately submit to evaluation by the instructor. Rarely, if ever, does the purchase of a product or service conform to this standard. In sum, the rights which students possess in the context of educational research do not follow from their role as consumers. That some students should not be favored in the classroom while others are disadvantaged follows, not from the fact that students have made equal payments, but from other principles concerning respect for individuals and the aims of the educational process. Fostering a respect for persons and, in turn, treating students as autonomous moral agents worthy of such respect are crucial elements of those aims. Were education to be provided free of charge, these rights would be guaranteed by virtue of those objectives.

These remarks indicate that claims about the aims of education can serve as a basis for making certain ethical judgments. An example of this is provided by Strike and Soltis (1985) in The Ethics of Teaching. In the following passage, these authors elaborate on what they see as an important educational objective:

In our view, growth as a moral agent, as someone who cares about others and is willing and able to accept responsibility for one’s self, is the compelling matter. Promoting this kind of development is what teachers ought to be fundamentally about, whatever else it is that they are about. We are first and foremost in the business of creating persons. It is our first duty to respect the dignity and value of our students and to help them to achieve their status as free, rational, and feeling moral agents.(17)

This goal is certainly laudable, but it is curious in one respect. There are no courses designed specifically to attain it. There is no course listed in the university catalog entitled “Achieving One’s Status as a Free, Rational, and Feeling Moral Agent.” If there is no course designed to fulfill such an important goal, how are students expected to achieve this status? The answer is that the goal is achieved by means of a variety of experiences which occur both inside and outside of the classroom. Student-teacher interaction is a crucial component of these experiences. Being treated with respect, as an end in oneself rather than as a means to something else, contributes to the development of a sense of worth and responsibility. Being treated as someone who is incapable of making rational decisions and who may legitimately be forced into a role extraneous to that of learner in the midst of one’s education is not conducive to the development of persons.

This line of reasoning makes explicit the link between claims about educational aims and particular ethical judgments. Nevertheless, it does not tell the whole story. Educational aims are diverse, and this diversity may lead to inconsistent implications for particular judgments. For example, education aims not only at producing persons in the sense intended by Strike and Soltis, but also at transmitting a cultural heritage and effectively imparting various elements of human knowledge. Emphasis upon this latter goal might lead to different judgments concerning the necessity for in-class experimentation and the appropriate roles of students and teachers in that context. One approach to avoiding such inconsistencies is to impose a hierarchy upon the diversity of educational aims. This is effectively what Strike and Soltis accomplish by their insistence that “we are first and foremost in the business of creating persons” (emphasis added). In any event, the point of these remarks is that statements of educational aims have a normative force and that they have logical implications for particular ethical judgments in education. In particular, the aim of creating persons is important and bears directly on the issues surrounding the use of controlled research in the classroom.

Back to the top

Go to: 3.3 Informed Consent and Student Obligation

Home > Teaching Resources > Computer Ethics Issues in Academic Computing > The Ethics of Evaluating Instructional Computing

The Research Center on Computing & Society
at Southern Connecticut State University
501 Crescent Street • New Haven, CT 06515
Director: (203) 392-6790 • e-mail: webmaster@computerethics.org

© 2000 – 2007 – Research Center on Computing & Society