Abstract
The present study was designed to determine which of three indices of sentence length most effectively predicted pre-schooler's correct sentence imitations : total number of words; number of content words only; or the density of content words within a sentence. Twenty-four pre-school children, mean age 3.11 years, were asked to imitate 32 sentences which varied in total number of words, number of content words, and lexical density. Analysis of the childrens' correct responses, using verbatim imitation criteria, revealed that the ordering of sentences obtained was one not predicted by any of the three length indices. Re-examination of the sentences suggested that the number of morphemes (the number of meaningful units), rather than the number of orthographically defined words, might be the appropriate counting index. A significant Spearman rank order correlation (r = + 0.67, 30 d.f., p <0.01) was found between morpheme length and sentence difficulty. It was concluded that morpheme length rather than word length may be a significant variable affecting pre-school children's sentence imitation performance.
Get full access to this article
View all access options for this article.