Knowledge Mapper Authoring System Prototype

We have developed a prototype authoring system for knowledge mapping that provides the capability to (a) specify map tasks (i.e., specify terms and links), (b) launch a knowledge map related to a task, (c) specify criterion or expert maps (for scoring purposes), (d) define groups of users and associate tasks to those groups, and (e) publish tasks for global access. The prototype interface is a simple HTML interface, with any destination function accessible within four mouse clicks. A usability study was conducted with one teacher and 62 fourthand fifth-grade students. Students were considered beta testers, and the teacher, who had used our existing knowledge mapper, was considered an expert. The teacher reported using the authoring system to create pretest and posttest tasks for students and to access and evaluate student maps. Students used the authoring system to create knowledge maps of their group research projects. The teacher reported that the authoring system was easy to use and thought student authoring promoted ownership of work. Students using the authoring system created more sophisticated knowledge maps, especially in the links specified by students. This anecdotal report is bolstered by analyses of the links used across all student-authored maps. Thirty-eight percent of the links were causal (e.g., causes) or functional (e.g., protects), and 24% were part-whole (e.g., part of). Limited screen real estate, confusing map access, complicated login functionality, and system crashes limited usability. Student performance on a usability task showed that, in general, students were able to independently carry out major authoring and mapping tasks; however, a positive relationship was found between the amount of time a student was at the keyboard and the amount of help that student needed during the usability task, suggesting that practice using the authoring system improves fluency with the authoring system. 1We wish to acknowledge and thank Joanne Michiuye and Linda de Vries of UCLA/CRESST for their help during data collection, and Ali Abedi of UCLA/CRESST for programming and technical support. We also wish to thank Terri Robinson for her content expertise. We also thank Jan Cohen, Margaret Heritage, and Sharon Sutton at UCLA’s Seeds University Elementary School (UES) for their assistance, participation, insights, and support. And finally we are deeply grateful to the creative and gifted teacher and her students at UES who participated and assisted in this research study.