PROCEDURAL NETWORK REPRESENTATIONS OF SEQUENTIAL DATA

Sequential data collected for usability testing, knowledge engineering , or cognitive task analysis are rich with information-so rich that in terpretation can often be overwhelming. This dilemma can be viewed as a data reduction problem. PRONET (PROcedural NETworks), a method for r educing sequential data in terms of procedural networks, is introduced and then applied and evaluated in two case studies-one involving huma n-computer interaction (HCI) in a simulated mission control operation at the National Aeronautics and Space Administration and the other inv olving avionics troubleshooting behavior for an intelligent tutor appl ication. The method involves five steps-collecting data, encoding data , generating transition matrices, conducting Pathfinder analysis, and interpreting procedural networks. The method employs the Pathfinder ne twork scaling algorithm, which is particularly suited for asymmetric d ata. Evidence is presented to support the descriptive and predictive u tility of this form of data reduction. In addition, lessons learned in applying PRONET to the two cases are discussed, applications of PRONE T to HCI are described, and guidelines are offered for using PRONET in exploratory sequential data analysis.