Date of Award:

5-2010

Document Type:

Dissertation

Degree Name:

Doctor of Philosophy (PhD)

Department:

Instructional Technology and Learning Sciences

Advisor/Chair:

Mimi M. Recker

Abstract

A chief human characteristic is the desire and ability to change the world. Prior planning is crucial when those changes are complex and extensive, and require the cooperation of many people. To satisfy this need, many disciplines have developed specialized notations for representing the plans. Developers in one discipline, computer-based instruction, are burdened by the current need to use two separate notations. Instructional experts design the instruction and represent the design with a primary representation. The instruction described in a primary representation is easy to see, which makes the representation suitable for evaluation, communication, and enhancement. Programmers translate the primary representation into a computer program, which is able to run on a computer but is a secondary representation.

The problem with this process is that the primary representation is equivocal or ambiguous. Equivocal representations are subject to multiple interpretations; it is also possible for programmers to introduce errors during translation. Alternatively, the computer program is unequivocal, but the instruction that is evident in the primary representation diffuses into the program, becoming obscure and difficult to use for further evaluation, communication, or enhancement. A representation that is both unequivocal and primary benefits computer-based instructional development by eliminating ambiguity and translation errors while preserving the instructional details for later use.

A representation is unequivocal if it is computable, and it is primary if it is able to represent the dynamic behaviors of complex instruction and its use as a design language can be demonstrated in published literature. My research evaluated and compared two design languages, PEAnets (networks of processes, entities, and actions) and the Unified Modeling Language, as potential unequivocal primary representations. Two translators, one for each language, were developed as a part of this research, and four complex computer-based instructional examples were created and translated into operational computer-based instruction. The translators demonstrated that both representations are computable, and the examples demonstrated that both languages are sufficiently robust to represent complex computer-based instructional systems. Both languages have been used successfully for designing instruction or general computer systems. I concluded, based on these observations, that both languages qualify as unequivocal primary representations.

Comments

This work made publicly available electronically on August 2, 2010.

Share

COinS