Information Technology Reference
In-Depth Information
1999). However, studies evaluating alternative approaches to training for end-user comput-
ing applications typically employ post-training tests to assess the comparative effectiveness
of various methods, such as lectures, self-paced manuals, and behavior modeling (Simon et
al., 1996) and exploration and instruction (Davis & Bostrom, 1993). Although this literature,
together with the computer programming literature, may offer suggestions for structuring the
presentation of information and problems to groups of learners, individual differences in such
factors as prior experience, motivation, and learning skills must be invoked to account for
learning outcomes that differ among the students. Alternatively, a focus on the individual
learner interacting with a tutoring system as a generic “teacher” might be anticipated to offer
benefits similar to those of a human tutor, where the objective is for the learner to attain a given
level of competency at the conclusion of a learning episode (Bloom, 1984).
Although guidelines to structure an entire course in Java have been published (e.g.,
Stiller & LeBlanc, 2001), it is the content that was emphasized, not the manner in which
students should be taught and assessed. In contrast, the present tutoring system design is
based upon principles that have yet to be applied and evaluated widely in computer-based
instructional systems having practical value in the classroom (cf. Anderson et al., 1995).
Rather than focusing on group outcomes as the only unit of analysis, the work reported in
this chapter first reveals the process of teaching and learning at the level of the individual
student. This is accomplished by analyses of the rate
of three learners' correct input,
response errors, and help requests on the tutoring system, all portrayed graphically as a
function of time in cumulative records. The purpose is to reveal the details of a learner's
disciplined study behavior in achieving mastery of the subject Java Applet. Next, we show
the application of the tutoring system to a class of students, and we present performance data
and ratings of the tutoring experience and self-efficacy.
LEARNING OBJECTIVES
The overall objective of the tutoring system is to teach a learner to construct a stream
of up to 36 Java items (i.e., atomic units, elements, items, or symbols) that together constitute
a Java computer program that displays a text string in a Netscape
©
browser window. The
approach taken is first to specify the learning objectives, which are as follows:
1.
Understand the meaning of each item of code as evidenced by passing a multiple-choice
test that is administered following the display of an explanation of the item.
2.
Enter the item into a key-in field by recall.
3.
Understand the meaning of one to several items that are presented in a row of code,
as evidenced by passing a multiple-choice test on each of 10 rows of code.
4.
Enter the row of items into a key-in field by recall.
5.
Enter the entire program into a text window by recall.
The objectives are accomplished by a learner's interactions with the tutor in a series
of incremental steps that progress to these objectives by the method of successive
approximation (Sulzer-Azaroff & Mayer, 1991). The outcome is the learner's correct produc-
tion of the program together with tested mastery of the meaning of each of the elements and
the interrelationships among those elements and similar mastery when the code is organized
into rows. Explanations of the meanings of the items and rows include several general rules
of object-oriented programming that are presented in the tutoring system.
