Some Effects of Computer Technology… - Background - The Research Center on Computing & Society

Text-only Version

Marvin J. Croy
Michael G. Green
James R. Cook

Terry Bynum opened the National Conference on Computing and Values with the question: “Shouldn’t computers serve to enhance and protect human values rather than to threaten them?” One feels compelled to answer this question in the affirmative. To do so, however, assumes that human values can be made explicit and that what counts as serving a particular value can be determined. In the following pages a description is given of a concerted effort to do just this. The effort takes place within an educational context in which computers are used for instructional purposes. There are two important characteristics of this attempt to assess the ways in which this computer use might affect human values. First, the assessment has both empirical and normative components. It endeavors to determine the empirical consequences of using computers in a certain way and to explicate the ethical value of those consequences. Second, this study aims at achieving an important goal in respect to the management of technology, namely, that of predicting and evaluating the consequences of introducing an innovation prior to actually implementing that technological change. At present the attempt to achieve this goal is only at the halfway point. We stand at the end of the first year of a two year study, and consequently the discussion here will focus not upon the particular empirical consequences emerging, but upon the techniques and rationale underlying the determination and evaluation of those consequences.

1. Background

The educational computer technology at issue here consists of computer assisted instruction programs for teaching deductive proof construction at The University of North Carolina at Charlotte. The development of these programs and indeed the development of much CAI in general has been stimulated by the promise of increased individualization. The promise has been to deal more sensitively and effectively with aptitude differences among students. These differences have been one of the chief difficulties faced in teaching deductive logic. Initially, two CAI programs were designed to provide the standard benefits of unlimited practice, self-pacing, and immediate feedback, but it was quickly seen that this was far from being sufficient. Student aptitudes and weaknesses varied so greatly that more needed to be done to provide genuine individualization. Much more needed to be known about the particular difficulties students were experiencing, and those difficulties needed to be categorized in ways that suggested helpful remedies. Consequently, the CAI programs were modified to function as data collection devices, thereby providing windows on student trouble spots. As students used the programs, records were kept of their rule application success rates, the particular errors made, and the steps taken in solving proof problems.

There have been three main effects of this data collection. First, by periodically generating class performance summaries, in-class instruction can be geared to current difficulties on a real-time basis. It turns out that different classes have somewhat unique characteristics and the CAI programs help in tracking and responding to widespread difficulties which may change as the semester progresses. Second, at the end of each semester, statistical analyses and pattern matches are used to identify more specific patterns of errors (“pseudo-rules”). The CAI programs have been modified to be sensitive to these patterns and tailor-made error handling routines have been built in.

The identification of pseudo-rules and other types of misconceptions, however, did not fully address the problem of individualization. There was a wide variety of ways in which students failed at proof construction, and there was no guarantee that any given misconception, however prominent in the data at large, would be troublesome for a given student. In addition, there was still much to be learned about student strengths and weaknesses. Consequently, special out-of-class meetings between student and instructor were initiated. These sessions have developed into an important part of the course. During these sessions, student and instructor sit together at the computer while consulting performance records and reviewing particular errors and problem solving efforts. What can result is the identification of idiosyncratic difficulties which are not widely shared among students. Occasionally, misconceptions are discovered which are not only unknown to the CAI programs but which have never before been encountered by the instructor. Overall then, the CAI programs provide both direct and indirect support for individualized instruction based on empirical inquiry into actual student difficulties. The data collecting features of the programs serve to (1) guide in-class instruction, (2) provide the empirical basis for investigating student errors, and (3) facilitate productive meetings between student and instructor.

The use of these student-teacher meetings in the effort to promote individualized instruction is of chief interest here. A number of questions concerning the impact of these meetings have arisen. While the impact upon student learning is certainly relevant, other questions appear to be of comparable or greater significance. These questions concern a wider effect on student behavior and attitudes. For example, introduction of the special student-teacher meetings appeared to coincide with some unanticipated yet beneficial changes in student behavior during class. Some students became more active in class, more inclined to ask questions and request clarifications. This is encouraging given the subject matter in deductive logic. There are few subjects in which one’s inability may be so clearly exposed to the class at large, and this prospect may serve to suppress classroom inquiries and responses.

While these events were promising, there was no clear indication that they were occurring consistently or that they were systematically related to the special out of class meetings. There was in fact no evidence that the special sessions achieved the goal of increased individualization or, even if achieved, that they did so in a more efficient manner than some other approach might. Given the number of hours required for the special meetings, the problem of efficiency seemed significant, but there was another issue that cast a critical shadow over these developments. The design of much current CAI runs counter to the direction in which this project has progressed, and this point bears elaboration in the context of “intelligent” programs and the developing impact of computers on education.

Back to the top

Go to: 2. Intelligent CAI and Individualization

Home > Teaching Resources > Computer Ethics Issues in Academic Computing > Some Effects of Computer Technology on Human Interaction and Individualization in the Teaching of Deductive Logic

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