Situated in San Antonio, Texas, Trinity University is a learning community committed to the highest standards of academic excellence. It has been ranked number one for ten consecutive years by U.S. News and World Report among universities in the Western United States. A recently implemented initiative, led by the Vice President for Information Technology, has set as its goal to enrich the education of Trinity students through the use of Internet-mediated learning technologies. As an experimental pilot study, testing the boundaries of the Internet-mediated learning environment, approval was granted to the author to offer his Computer Science course on the Principles of Data Abstraction, Fall Semester 2001, in the model of the virtual university (no formal classes; all learning taking place via the Internet). This paper describes the parameters of this study, outlining the specific course objectives and requirements, with emphasis on the unique pedagogical features of the course design. A comprehensive evaluation of the study is included.
The object of study is a modern, laboratory-supplemented course in the Principles of Data Abstraction, as implemented in the C++ programming language, with emphasis on the notion of an abstract data type (ADT). There are two texts [1,2], the latter consisting of nine experiments having titles as shown:
Accordingly, the course is divided into nine week-and-a-half periods, during which an individual experiment is discussed (and performed by the students), as a supplement to readings in the main text [1].
The course grade is determined by a student's performance on two in class examinations (one midway through, the other at the end of the semester), nine laboratory quizzes (submitted by email at the beginning of a laboratory period), six-of-nine laboratory reports (submitted hard-copy to the instructor's real-world mail box at the end of a laboratory period), and participation in the bulletin board (See details below). The quizzes, routine questions on the laboratory experiment and related readings, are emailed to all students simultaneously as an addenda to the comprehensive study guides (Again, see details below). And as suggested above, each student is required to perform six experiments (of the nine that are studied), and to write formal reports on these six investigations.
Besides the previously stated requirements of a student, he or she is also expected to attend three scheduled in class evaluations of the Internet-mediated environment, where the instructor has the opportunity to determine the effectiveness of the unique pedagogical course design presented here, from a student perspective. Student responses to detailed questionnaires at these "get-togethers" constitute one focus of the end-of-semester evaluation that the instructor has prepared. Except for these meetings, all contact with the instructor is via the Internet (or at his scheduled office hours, when problems of understanding are so acute that they cannot be resolved in any other way).
A critical component of the Internet-mediated environment designed here is the bulletin board, an email forum for discussion of the course material. Two or three days before the beginning of each of the nine laboratory experiment periods, the instructor broadcasts to all students simultaneously, a study guide, assigning, outlining, and commenting on the related readings (particularly as concerns the laboratory experiment at hand), pointing out "trouble spots," etc., and this email message serves as a focus for the discussions to follow. Students are encouraged to post questions, comments, etc., relating to the topic(s) under discussion, after which other students will post answers, further questions, comments, etc., in a continuing dialogue (the instructor will also participate, in the role of moderator and tutor). Each student posting is broadcast to the class as a whole, after perhaps some editing by the moderator, though he or she only communicates directly with the instructor. In this way, the bulletin board functions much like an Internet moderated mailing list.
As a major thrust of the recently implemented Trinity University Internet-mediated teaching environment, the Blackboard 4 e-learning System has been installed. It is of interest to compare the blackboard and the bulletin board methodologies in relation to the course at hand. With the blackboard, an ongoing dialogue can again be established, but as an informal, freely accessible media, open to any student who wishes to express himself or herself (hopefully in a manner that has some bearing on the course at hand). In its functionality, the blackboard is more like an Internet chat room. The instructor looses all control over the discussion. Such a free-flowing medium may be "just the ticket" in certain seminar courses, where individual opinions are admitted, even welcomed. But in a technical course, such as that under consideration here, an unmoderated, unmodulated posting of a student might be confusing and misleading to those others in the class who are tuned in to the blackboard. With the bulletin board methodology, as described above, the instructor has ultimate control over the content of the discussion. Often students in computer science are not able to phrase their questions or comments in the proper terminology. With the bulletin board, student questions can be rephrased by the instructor-moderator for technical accuracy, so as to ensure that the proper understanding is being conveyed to the rest of the class. Each technology undoubtedly has its place. But our experience thus far has provided overwhelming evidence that for the course under study, the right choice has been made between these two competing methodologies.
By every measure that the instructor could apply to the performance of students in the experimental pilot course offering, as opposed to that of students of previous years in the more traditional classroom setting, results were seen as favorable. The median for he midterm exam was within a percentage point of that from previous years. The same held true for the final examination. And laboratory reports, if anything, were of a superior quality.
However, it is in the transcriptions from the bulletin board that the instructor draws his most favorable impressions. These postings, some seventy pages of text, represent the most detailed and substantive interactions that this instructor has ever had with a group of students. In the traditional classroom setting, many students are reluctant to ask a question or to participate in discussions, whereas with the online bulletin board, somehow these same students are motivated to become involved. And the instructor, rather than giving offhand and perhaps only half-correct answers to student questions, as we often find in the classroom setting, has the time and the motivation for giving the clearest and most precise responses. All in all, the idea of the bulletin board is seen as the most successful component of the Internet-mediated environment designed here. One could argue for its inclusion as a supplement to the course delivery system, whatever the teaching environment.
But the real test is to be found in the student evaluation of the course. In a questionnaire submitted to the students at the end of the semester, the following are representative of the responses:
Only 3 out of 20 students indicated that, knowing what they know now (at the end of the semester), they would "prefer to have taken the course in a traditional classroom setting." Even when an individual student may have been somewhat critical of the Internet-mediated environment, they were good-humored in their response:
But all in all, the student reaction to this experimental course offering has been most encouraging and supportive. The class was especially pleased to learn that a paper describing their experience had been accepted for publication at this international conference. And it is my pleasure to report the results of this study in this forum.
1. F.M. Carrano, Data Abstraction and Problem Solving in C++, Third Edition, Addison-Wesley, 2001.
2. R.E. Prather, Laboratory Manual for Data Abstraction, Trinity University (http://www.cs.trinity.edu/~rprather), 2001.