This paper outlines a study exploring the language of science textbooks at the transition from primary to secondary education.  The study is focused on textbooks in English for native speakers (in Ireland and the UK) and for non-native speakers learning in a CLIL (Content and Language Integrated Learning) setting.  The two central research questions are: 

  is there a qualitative difference in the language of science textbooks at late primary and early secondary level?

  is this difference consistent across different first language (UK and Ireland) and second language contexts (CLIL settings)? 

This study forms part of a larger project to explore the language of science at transition points in education in textbooks and in classroom discourse and pupils’ reactions at these transition points.  A recent study on science at primary level in Ireland by the National Council for Curriculum and Assessment identified “the perceived difficulty of post-primary school science” (NCCA, 2009, p.17) as one of the key factors contributing to a decline in positive attitudes towards science.  This research will investigate to what extent language contributes to the perceived difficulty and how to tackle and overcome this linguistic barrier and impact positively on student learning outcomes.  The first stage in this work is to explore the language of textbooks with the aim of characterising the features of this discourse.  This paper describes a comparative corpus analysis of textbooks before and after the transition to secondary education in different European contexts.  The objective is to characterise features of textbooks at primary level and at secondary level and explore any substantive distinctions between them in terms of discourse in order to identify possible areas of difficulty in processing for pupils.  The comparison of first and second language contexts is intended to address the issue of heightened sensitivity to language issues and whether this impacts on discourse features. 

The discourse focus draws on work on the discourse of science and science education (Halliday and Martin (1993), Martin and Veel (1998)).  The analysis is conducted within the framework of lexical cohesion (Halliday and Hasan (1976), Hoey (1991)) and uses computational methods to provide an objective analysis of text structure and content upon which to base comparisons.  The findings may have substantial implications for science educators and/or publishers where the transition from primary to secondary must be bridged linguistically as well as conceptually. 

Devitt, A. and K. Ahmad. (2007). Cohesion-based Sentiment Polarity Identification in Financial News. In Proc. of 45th Annual Meeting of the Association for Computational Linguistics (ACL’07), Prague, Czech Republic, June 2007. Devitt, A. (2004). Methods for Meaningful Text Representation and Comparison, PhD. Thesis, Department of Computer Science, Trinity College Dublin. Grabe, W. and F. L. Stoller (2002). Teaching and Researching Reading. London, Pearson Education Longman. Halliday, M.A.K; and Ruqayia Hasan (1976). Cohesion in English. London: Longman. Halliday, M. and J. Martin (1993) Writing Science: Literacy and discursive power. London: Falmer Press. Hoey, M. (1991): Patterns of Lexis in Text. Oxford: OUP. Kamil, M., J. Mosenthal, et al. (2000). Handbook of Reading Research, Volume III. Hillsdale, NJ, Lawrence Erlbaum Associates. Martin, J.R. and Veel, Robert (eds.) (1998) Reading Science Critical and functional perspectives on discourses of science. London and New York: Routledge. National Council for Curriculum and Assessment (2009) Science in Primary Schools, Phase 2: Final Report. Available at: http://www.ncca.ie/uploadedfiles/Primary/Science_Phase2_Final_report.pdf [Accessed on 20th January 2011].