Accessibility navigation


An investigation into corpus-based learning about language in the primary school

ESRC Research Award Ref. R000223900

Award holders:

Alison Sealey, Paul Thompson (University of Reading) and
Mike Scott
(University of Liverpool)

Amount: £39,578

Duration of project: 01 June 2002 to 31 May 2004.

The End of Award report was graded 'Outstanding'
by the ESRC.
(November 2004)

This project investigated the potential of corpus-based teaching for children's evidence-based learning about language. Corpus-based teaching involves the use of:

  • large, electronically stored collections of authentic language
  • an interface of computerised tools for identifying and describing its features
  • activities designed to direct learners' attention to those features.

A sub-corpus was constructed, from the British National Corpus, of texts written for children. One strand of the research involved a linguistic analysis of this specialised corpus in comparison with reference corpora.

The research also investigated how children at Key Stage 2 (the upper primary age-range) respond to language presented in this format. Teaching materials and activities which draw learners' attention to features of grammar and vocabulary were devised from the corpus, and a series of teaching sessions was carried out with groups of children from two schools, one from a Year 4 class (of 8-9 year olds) and the other from a Year 5 class (9-10 year olds). The children were observed as they completed the tasks set and devised investigations for themselves. They were also interviewed about the concepts involved.

Based on the first phase of fieldwork, a new interface appropriate for primary school pupils was developed. Such interfaces allow the user to identify and demonstrate patterns in grammar and vocabulary, and modifications to an existing corpus analysis program, Wordsmith Tools, were made to support these young learners in their own linguistic investigations. A second phase of fieldwork was conducted, with some of the original children and some 'novices', involving more direct interaction with the corpus itself.

  • To see a list of presentations and publications arising from the project, click here.
  • For a Powerpoint presentation outlining the classroom-based work and initial findings, click here. (Best viewed in Internet Explorer)
  • For a Powerpoint presentation which explains what we have discovered about the distinctive features of the language of the CLLIP corpus, click here. (Best viewed in Internet Explorer)

Things to do

Contact Us

Tel: 0118 3788141

Email: appling@

Page navigation


Search Form