Abstract Decomposition of Spelling Errors
- 177 Downloads
There has, hitherto, been no adequate method to found a taxonomy of spelling errors based on any substantial cognitive theory of spelling behavior, because there has been no theory adequate for the purpose. But such a taxonomy is needed. This paper begins, as it were, at the other end, by creating an ‘abstract’ taxonomy that might lead to an adequate theory. Suppose the spelling of a word to be determined by a collection of abstract conditions. If a subject satisfies all the relevant conditions, the word is spelled correctly. If two subjects satisfy an identical pattern of conditions they spell the word identically (though the converse need not be true).
Formal development leads to an algorithm for constructing a minimal set of conditions to represent formally a subjects x words arrangement of spellings. The algorithm is applied to the spellings of 109 words from two groups (‘dyslectics’ and ‘normals’) of 26 schoolchildren. Assumptions are sufficiently elementary that they do not invoke any semantic link to spelling errors as such. Consequently, the reasoning and the algorithm ought to have wide applicability.
Unable to display preview. Download preview PDF.
- Azmoodeh, M. (1988). Abstract data types and algorithms. Houndmills & London: McMillan.Google Scholar
- Grimaldi, G. (1989). Discrete and combinatorial mathematics (2nd ed.). Reading etc.: Addison Wesley.Google Scholar
- Smolenaars, A. J. (1989). An elementary formal categorization of a corpus of spelling errors. In E.E. Roskam (Ed.), Mathematical Psychology in Progress (pp. 197–213). Berlin, etc.: Springer Verlag.Google Scholar
- Smolenaars, A. J. (1990). A sequel to an elementary formal categorization of a corpus of spelling errors. Unpublished manuscript. Report nr. 114 of the project ‘Knowledge Acquisition in Formal Domains.’ Amsterdam: Department of Psychology, University of Amsterdam.Google Scholar