Sixth International Conference on Immersion and Dual Language Education: Connecting Research and Practice Across Contexts

Paper/Best Practice Session (1 hour)

Minority language education (FL1) and French immersion education (FL2)

One-Way Second/Foreign Language Immersion

A summary should be a coherent, accurate, and succinct representation of a source text. Our analysis of summary writing by Canadian students in both minority Francophone (FL1) and French-immersion (FL2) programs from Grade 9 to university level will focus on variables related to grammatical proficiency, syntactic complexity, and connector use.

A summary should be a coherent, accurate, and succinct representation of a source text. Beyond knowledge of the topic, writing a good summary is predicated on many interconnected skills underlying effective reading and writing. As student language proficiencies may differentiate the quality of summary produced, we hypothesize that both language program and grade or academic level of participants may be important determinants.

We will present statistical analyses of grammatical proficiency, syntactic complexity, and connector use in summaries produced by both Francophone (FL1) and French-immersion students (FL2) in Manitoba. This cross-age study is based on two corpora. The first corpus includes 400 summaries written by students across grades nine to twelve, 50 in each grade and language program, in response to a French-language text dealing with the re-introduction of wolves in Yellowstone Park. AUTHOR and his colleagues (2001, 2004, 2007) analyzed both content and language variables in this corpus and reported many statistically significant differences for program and/or grade, as well as interactions between these categorical variables. The second corpus includes 87 additional summaries produced by first-year university students based on the same text and similar procedures. The results of this latter study have not yet been published.   

For the 37 grammatical variables, program-grade interactions were observed for four of the six categories of errors: noun, syntax, verb, and spelling. The remaining two categories (expression and pronoun) showed significant differences for the effect of program only. A fine-grained analysis will be presented for variables of particular interest, as well as those related to syntactic complexity and connector use.