Determination of bilingualism subtypes and their relationship with linguistic abilities in Lebanese bilinguals

Abstract

Aims and objectives: Profiles of bilinguals vary among studies due to the diversity of intrinsic and extrinsic factors, bilingualism classification and to the discrepancies between measures. Thus, a generalizable index with a defined threshold is needed to capture the linguistic dominance of bilinguals and facilitate comparisons between studies. This study’s objective was to define and examine the validity of a new bilingualism index. Methodology: This index was derived from the Language Experience and Proficiency Questionnaire (LEAP-Q) conducted on 100 bilingual Lebanese participants (age = 68.2 ± 9.4 years; education = 12.8 ± 5.2 years). Data and analysis: The bilingualism index was based on selected items of the LEAP-Q, optimizing the explained variance on a linear regression using the differential Arabic-French score on the Boston Naming Test as the dependent variable. The validity of the classification was examined using other linguistic (articulatory rate, shortened Token test, Stroop reading subtest) and cognitive screening tests (mini-mental state examination [MMSE]). Findings and conclusion: LEAP speaking and oral comprehension scores provided a parsimonious index that accounted for naming variance ( R2 = .435, p = .0001) and subdivide our population into three bilingualism subgroups (prominent Arabic, balanced, prominent French). A prominent language advantage was found in the expected direction on linguistic (articulatory rate: p = .03; shortened Token test: p = .026; Stroop reading subtest: p = .0001) and cognitive screening tests (MMSE: p = .08). Originality and implications: These results show that a simple index can accurately characterize adult bilingualism subtypes and offers clinicians an easy and fast tool compared with the usual procedure used to determine individuals and patient’s bilingualism subtype. More broadly, the index’s validity in other bilingual populations is warranted for generalizability of the present findings.