Some Effects of Computer Technology… - First Year Objectives of the Study - The Research Center on Computing & Society

Text-only Version

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

3. First Year Objectives of the Study

1) The primary objective of the first year of this study was to develop a set of measures and techniques which would be sensitive to differences resulting from students receiving human as opposed to computer supplied feedback. In particular these measures were to include:

a) a questionnaire for assessing student attitudes toward the course, instructor, computers, subject matter, and fellow classmates;

b) variables which identify relevant in-class behaviors; and

c) performance measures which would assess any impact on student learning

2) Two sections of PHI – 2105 (Deductive Logic) were to be involved in a within-subjects experimental design in the process of developing these measures.

During the Fall semester of 1990, twenty six students enrolled in PHI – 2105 signed informed consent statements and were divided into two groups. These groups were formed as a result of matching subjects on the basis of previous computer experience. Students provided information about their prior computer experience as part of a logical ability pre-test. Early in the course, students also filled out a questionnaire (of approximately 35 items) which was subsequently repeated three times during the remainder of the course. This initial phase of the course is referred to as time frame 1 as shown in figure 1. (Figures are included below at the end.) During this period students were assigned seats for the duration of the course in order to facilitate documentation of in-class behavior. The procedure of video taping classes was also introduced. Most classes followed a lecture/ demonstration format as the initial concepts and procedures of symbolic deductive logic were established.

This period was followed by time frame 2 which marked the first crucial stage of the study. Students took up the task of learning deductive proof construction and began using two CAI programs designed to facilitate that effort. At this point the class was divided into the two groups which were matched in terms of previous computer experience. Students designated as members of group A attended two special out of class meetings (approximately 20 – 25 minutes in length) with their instructor. During these sessions, the instructor made a diagnosis of current difficulties and a recommendation for addressing these problems. This diagnosis was recorded on a form designed for this purpose. Group B students meanwhile were receiving their diagnostic feedback directly from the computer. The same form filled out during the group A teacher-student meetings was available for group B students as a screen display in one of the CAI programs. Although this process was characterized for students as “receiving feedback from the computer,” the diagnostic feedback was actually entered on a case-by-case basis for each student by the instructor. The instructor generated this feedback by examining records of student performance in exactly the same way as was done during the teacher-student meetings. Consequently, the content of the feedback was very similar, if not identical, for particular observed difficulties whether the student having those difficulties was in group A or in group B.

Classroom instruction during this period was centered around problem-solving activities. For example, a proof problem would be written on the board and students would be given several minutes to individually search for a solution. Then students would describe alternative solutions as the instructor asked questions about the appropriateness of the proposed solutions and particularly about the strategies used in discovering those solutions. In addition, some students usually asked their own questions about the proofs described. This pattern of in-class activity prevailed for much of time

At the end of time frame 2 all students faced an hour long in-class exam emphasizing deductive proof construction. Subsequent to that exam, students filled out a second questionnaire which was an adaptation of that previously given. Two forms of this questionnaire were administered. One form contained elements which asked students in group A specifically about their special meetings with the instructor while the second did the same for group B students concerning their computer supplied diagnosis. Both forms of this questionnaire asked students about their attitudes toward the instructor, their classmates, the course, computers, and about their current performance.

During the next phase of the course (time frame 3), the activities of students in each group were switched. Group A students now received their diagnostic feedback from the computer while group B students began meeting individually with the instructor. The subject matter remained deductive proof construction (although the complexity and level of difficulty increased) and in-class procedures continued the problem solving patterns of time frame 2. Records of student attendance and in-class responsiveness were kept as in time frame 2. At the end of this period, students once again faced an hour-long in-class exam and filled out a questionnaire relevant to their group’s activities.

In the final phase of the course (time frame 4) the subject matter shifted toward the application of deductive logic to natural language argument reconstruction. No CAI programs were used during this period, nor were any special student-teacher meetings held. Although this was not a crucial period for research purposes, the monitoring of student attendance and responsiveness was continued. (Classroom procedures were similar to those of time frame 2 and 3.) At the end of the course, students once again completed a one-hour exam and a final questionnaire. The questionnaire was adapted toward explicit comparisons of the treatments earlier experienced by students.

During the Spring 1991 semester, 29 students enrolled in PHI – 2105 were also studied using the within subjects design. The general structure of the course and pattern of activities were identical to those of the Fall semester. The informed consent statement and relevant portions of the exams also remained identical. However, the questionnaires were refined on the basis of previous findings.

4. First Year Results

Prior to presenting actual results, it should be reemphasized that an important component of this study focuses not upon how human versus computer-supplied feedback affects student learning, but rather how this difference affects human interaction and student attitudes. In particular, student activity in the classroom and interaction with the instructor is of interest. This interest grows out of previous experience with the special student-teacher meetings. The introduction of these meetings emerged from the long term development and use of supplementary CAI programs. These programs evolved in a form which provided students with rudimentary information concerning their individual strengths and weaknesses. Eventually, special student-teacher meetings were built around examinations of performance records assembled during the use of the CAI programs. It was at this point that casual observations seemed to indicate that students became more active and responsive in class following these meetings. There seemed to be something about the human contact itself that produced changes in student in-class behavior, and these changes seemed to carry over in lesser degrees to subsequent parts of the course. This supposition, however, was untested prior to our current research efforts.

Back to the top

Go to: 4.1 Predictions and Outcomes

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