On-Line Assessment of Expertise Project
Quarterly progress rept. Jan-Mar 1992.
NAVAL UNIT FORT DETRICK FREDERICK MD
Pagination or Media Count:
The initial study was designed to test the user interface for gathering data about problem solving, test our choice of problems, and gather verbal protocols that could be compared to the user-interface action sequences collected by the system. Ten subjects were drawn from the Pitt into physics class for science and engineering majors. Each subject was first introduced to the system and taught how to use the editors to solve problems. They were also shown how to call up any of the ten example from their textbooks and read it via the poor-mans eyetracker. They then solved problems relevant to the target chapter chapter 5 -- Newtons laws applied to translational motion without friction but taken from other textbooks. They worked for two hours, solving as many problems as they could. They were asked to give a verbal protocol as they worked, and prompted when they fell silent. The first five subjects uncovered significant problems with the user interface, so their runs were truncated andor protocols were not taken. They helped us improve the interface to the point where subjects now find it simple to learn and use, but their data has been discarded. Subsequent analysis will be based on the five remaining subjects.
- Computer Systems
- Human Factors Engineering and Man Machine Systems