The role of age and timing in bilingual assessment: non-word repetition, subject-verb agreement and case marking in L1 and eL2 children with and without SLI

ABSTRACT

Diagnostics in bilingual children is challenging, due to an overlap of production patterns in typically developing (TD) bilingual children and monolingual children with specific language impairment (SLI). To screen bilingual children effectively, the Language Impairment Testing in Multilingual Settings (LITMUS) tools were developed in an international project. The present study tests three of these tools for their suitability and diagnostic accuracy for early second language learners (eL2) of German, aged six to eight years. The study focuses on the timing in first language (L1) TD acquisition, investigating early and late acquisition phenomena of the morphosyntactic domain (subject-verb agreement [SVA], and case marking), combined with a non-word repetition (NWR) task targeting phonological complexity. The study aims at evaluating these three LITMUS-tools regarding their diagnostic accuracy, compared to a standardised assessment tool (LiSe-DaZ).

To this end, forty-two children were tested using the LITMUS-tools, namely, contrastive case marking (CCM), supplemented by an elicitation task for the prepositional case, SVA and NWR. Four groups of children participated: eL2 children with SLI (mean age 7;6, mean age of onset 3;1), eL2 children with TD (mean age 7;10, mean age of onset 2;11), L1 TD children (mean age 7;3) and L1 SLI children (mean age 7;2). Results show NWR and SVA as suitable markers and the LITMUS-tools as suitable screenings. Conversely, CCM does not disentangle SLI from TD in the investigated bilingual population by this age.

KEYWORDS:

• Developmental language disorder
• bilingualism
• German as a second language
• case marking
• subject-verb agreement
• non-word repetition
• screening tools; specific language impairment

Acknowledgments

The author thanks Lena Kliemke for her help with data collection and the Equal opportunity office of the University of Hildesheimfor funding a student assistant.

Disclosure Statement

The author reports no conflict of interest.

Notes

¹ For a discussion of the narrow term SLI and the broader term DLD (developmental language disorder), see Bishop (2017). We use the term SLI for this study in order to describe our well-defined sample of children with an impairment in language acquisition with no obvious cause and an absence of cognitive impairments (Leonard, 2014).

² For a theoretical explanation within the generative paradigm for the order of acquisition and the relation between AoO-effects, input and timing, see Tsimpli (2014) and Schulz and Grimm (2019).

³ For detailed information about LITMUS-NWR-German, see Abed Ibrahim and Fekete (2019), Grimm and Hübner (in press), and Schulz and Grimm (2020).

⁴ Due to institutional concerns regarding lengthy background questionnaires for parents, we were forced to shorten the PaBiQ. Therefore, we do not have comprehensive background information for all children.

⁵ The LiSe-DaZ (Schulz & Tracy, 2011) is normed for children up to seven and up to a LoE of 71 months. None of the participating eight-year-old children exceeded the LoE of 71 months. They were tested by means of LoE values of the LiSe-DaZ.

⁶ In the BiSLI group, one child had to be excluded from the SVA analysis because task demands regarding perspective switch were too high.