An inquiry of time and cost estimating for computer-based training courseware design and development as determined by the modified Delphi method
Abstract
This study investigated the approaches used to quantify "one hour" of CBT and identified the average level of effort and cost estimates for CBT design and development. Key factors that affect CBT time and cost estimation, and several estimating guidelines were identified. A modified Delphi Method was used involving 25 CBT experts from 16 organizations between March and May of 1991. The study showed that the concept of "one hour" is used to quantify the amount of CBT by most experts. Three methods used as units of measure to quantify one hour of CBT were identified: the number of objectives, the number of screens, and the number of interactions. Seventy-six percent of the CBT experts surveyed indicated that they use between three and eight simple objectives to quantify one hour of CBT. More than one-half of the CBT experts indicated that they use between 76 and 125 screens to quantify one hour of CBT. Even though 75 percent of the experts surveyed attempted to quantify one hour of CBT in terms of the number of interactions, there was no consensus as to how many interactions constitute one hour of CBT. The average level of effort estimated to develop the first, second, fifth, tenth, fifteenth, and twentieth hour of CBT were 240, 190, 160, 140, 130, and 120 hours respectively. Based on this data, a mathematical equation was derived to approximate the hours to develop each hour of CBT. The average number of hours to develop one hour of CBT for mainframe CBT, microcomputer CBT with limited graphics, and microcomputer CBT with extensive graphics were 160, 180, and 230 hours per hour respectively. Twenty-eight factors considered when estimating CBT time and costs were identified and ranked in order of the significance of their impact when developing estimates. The top rated factors were: interface complexity and complexity of branching structure, complexity of response analysis, level of interactivity, the skills and experience of the project team in instructional design and CBT, the type of authoring system used, the complexity and stability of subject matter, and number of client review and revision cycles.
Degree
Ph.D.
Advisors
Lehman, Purdue University.
Subject Area
Curricula|Teaching|Computer science|Educational software
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.