Language preference in citations: a quantitative analysis of MA theses at a French-speaking university

ABSTRACT

Pluriliterate academic practices offer students who are literate in both their L1 and English the possibility of relying on sources in multiple languages. Surprisingly, there is little research on the factors playing a role in students’ choice of sources and whether this choice is related solely or partly to the language of the source. This study examines whether students from a French-speaking Belgian university use pluriliterate strategies when writing their master thesis (n = 240). The objective is to investigate to which extent the selection of sources written in English rather than in other languages depends on the language of writing and on the faculty where the thesis was submitted. This study shows that students writing in English tend to have a more pronounced preference for sources in English in their readings, with estimated proportions above 93.6% across faculties, compared with students writing in French on fairly similar topics, especially in the social sciences. For example, theses in Management have an estimated proportion of sources in English between 90.1% and 96.5% when written in English and between 35% and 61.1% when written in French. These findings highlight the need for investigations on students’ attitudes towards pluriliteracy.

KEYWORDS:

• Biliteracy
• pluriliteracy
• academic writing
• citation
• multilingualism

Acknowledgments

This research has benefitted from the statistical consult with Statistical Methodology and Computing Service, technological platform at UCLouvain – SMCS/LIDAM, UCLouvain.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 The GLMM is an extension of the linear mixed models, or linear regression model, to cases where the variable under study is categorical rather than continuous (Jiang, 2007, p. 119).

2 As international bibliographies are biased towards English sources, Yitzhaki (1998) suggests a correction: the estimated proportion of works in other languages is multiplied by 2. In 1999, Bookstein and Yitzhaki propose yet another more complex correction (Bookstein & Yitzhaki, 1999). We refer to the latter article for more information on the subject.

3 As noted by Lillis and Curry (2010, p. 7), the term ‘international’ is ambivalent. In this context, ‘international’ is often used to refer to Anglophone-centre-based publications.

4 Here, the expression ‘international topics’ refers to topics in which more than one national or linguistic context is studied.