ERIC Number: ED235202
Record Type: RIE
Publication Date: 1981-Apr
Finding "Problem Types" in Judgments of Problem-Similarity: Comparison of Cluster Analysis with Subject Protocols.
Herring, Richard D.
Literature in mathematic problem-solving suggests that learners store information in memory which helps them solve stereotyped algebra word problems. Cluster analysis has been used as an exploratory tool to infer the types of problems which have common representations in memory. This study compares the results of a hierarchical cluster analysis of judgments of problem similarity, with the open-ended rationales that subjects give for their similarity judgments. Results indicate that interpretation of a hierarchical cluster analysis is highly dependent upon the analyst's criterion of similarity. Evidence in support of two conflicting hypotheses can be taken from the same output. A critical assumption of the Johnson method of hierarchical clustering is that the data satisfy the"ultrametric inequality," and it is found that this assumption is more stringent than is commonly realized. (Author)
Publication Type: Speeches/Meeting Papers; Reports - Research
Education Level: N/A
Authoring Institution: N/A
Note: Paper presented at the Annual Meeting of the American Educational Research Association 65th, Los Angeles, CA, April 13-17, 1981).