Creating a Story-Based Dynamic Assessment to Identify Developmental Language Disorder in Children Learning English as an Additional Language

Abstract

Background: Developmental Language Disorder (DLD) hinders the language acquisition of around 8% of UK children at school entry, including those with English as an Additional Language (EAL). Early diagnosis is essential for mitigating its impact on everyday learning and communication. However, detecting DLD in multilingual children is difficult due to the absence of suitable assessment tools and practitioners fluent in the children’s home languages. This exacerbates the challenge of distinguishing between language issues stemming from DLD and from a limited familiarity with English. In this context, Dynamic Assessment (DA) emerges as a solution that integrates teaching with assessment to uncover a child’s capacity to learn language, rather than their current language skills, reducing linguistic and cultural bias and catering to children with variable English abilities. This work focuses on creating a DA designed to assist DLD detection in UK school-aged children with EAL from diverse home language backgrounds. Using storytelling in English, this DA targets learning potential across three DLD-vulnerable areas: narrative macrostructure (story grammar and episodic structure complexity), emotional vocabulary, and receptive affective prosody. Methods: Following pilot studies, the DA was trialled with 14 children with EAL aged 4;06– 8;11 years from Northeast England. Its effectiveness was tested by comparing children’s DA performances (scores in receptive affective prosody and story generation tasks, and modifiability) relative to outcomes in measures relevant for DLD diagnosis: the Crosslinguistic Nonword Repetition Test (CL-NWRT), New Reynell Developmental Language Scales (NRDLS) in English, and variables concerning English experience, proficiency, and presence of DLD risk factors. Results: In children’s narratives, story grammar usage significantly improved over the DA’s teaching phase, but episodic structure complexity and emotional vocabulary did not, nor did affective prosody understanding. Correlational and predictive relationships between DA performances and NRDLS scores, along with English proficiency, highlight the need to refine the DA to assess learning potential irrespective of existing English language skills. Such relationships with the CL NWRT and DLD risk factor scores support the DA’s sensitivity to DLD-related vulnerabilities and value in identifying DLD among children with EAL. Conclusions: Additional testing with a larger sample is essential to strengthening the findings, which call for refinement of the DA to better evaluate DLD risk across UK multilingual children with varied English language experience and proficiency.