THE MDS-ANAVA TECHNIQUE FOR ASSESSING KNOWLEDGE REPRESENTATION DIFFERENCES BETWEEN SKILL GROUPS

Knowledge representation is one of important factors that determine hu man performance on cognitive tasks. Due to different levels of experie nce, different groups of people may develop different knowledge repres entations which lead to different levels of performance on cognitive t asks. If knowledge representation differences exist between skill grou ps such as experts and novices, those differences can be used to guide the training of novices for skill acquisition, and to assist the desi gn of jobs and tools for performance enhancement. A technique is prese nted in this paper for assessing knowledge representation differences between skill groups, based on multidimensional scaling (MDS) of dissi milarity data and analysis of angular variance (ANAVA). The MDS-ANAVA technique was applied to two sets of dissimilarity data that were obta ined from ten experts and ten no,ices in the computer domain, one set concerning 23 concepts in C computer programming, and another set conc erning 21 concepts in the UNIX operating system. knowledge representat ion differences from the MDS-ANAVA technique are compared with those f rom the hierarchical clustering technique. The MDS-ANAVA technique sho ws several advantages to the hierarchical clustering technique in test ing the statistical significance of knowledge representation differenc es between skill groups and revealing features underlying knowledge re presentations of skill groups.