Assessment of complement clauses

A comparison between elicitation tasks and language sample data

Gillian Steel*, Miranda Rose, Patricia Eadie, Rosalind Thornton

*Corresponding author for this work

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

The aim of this study was to determine whether data from specifically designed elicitation tasks and commonly-used language sampling techniques provided the same information about children's production of complement clauses. It was predicted that elicitation tasks would yield more examples of complement clauses and children would use a wider range of verbs to form complement clauses in elicitation tasks than in language samples. A group of 20 pre-school children aged between 3;11-5;3 years were investigated. Each child completed two elicitation tasks, prompting the production of two of the major forms of complement clauses. A language sample of 100 utterances was also collected from each child. The results demonstrated that the two methods of data collection do not provide the same information about children's production of complement clauses. Significantly more examples of both types of complement clauses investigated were produced in the elicitation tasks. Similarly, children used a greater range of verbs in the elicitation tasks. Thus, language sample data under-estimate pre-school children's competency with complement clauses. Accurate and efficient methods of assessment of children's language structures are vital for speech-language pathologists to base their management decisions on. Elicitation tasks offer a viable alternative to language sampling.

Original languageEnglish
Pages (from-to)286-295
Number of pages10
JournalInternational Journal of Speech-Language Pathology
Volume15
Issue number3
DOIs
Publication statusPublished - Jun 2013

Fingerprint Dive into the research topics of 'Assessment of complement clauses: A comparison between elicitation tasks and language sample data'. Together they form a unique fingerprint.

  • Cite this