| Author: | Hamdani, Saboor Zafar |
| Title: | Assessing Urdu-speaking multilingual children in Hong Kong and Pakistan : heritage language acquisition and developmental language disorder |
| Advisors: | Chan, Angel (CBS) |
| Degree: | Ph.D. |
| Year: | 2024 |
| Department: | Department of Chinese and Bilingual Studies |
| Pages: | xx, 253 pages : color illustrations |
| Language: | English |
| Abstract: | This dissertation examined the language abilities of children acquiring Urdu as their first language (L1) in Hong Kong and Pakistan, addressing a range of conceptual and methodological issues in multilingual child language acquisition. Study 1 addressed two key issues in the heritage language acquisition of multilingual ethnic minority (EM) children: the effects of reduced language experience and linguistic vulnerabilities. This study assessed a group of typically developing (TD) EM multilingual children acquiring Urdu as L1 and minority heritage language (The EM group; N=31; mean age = 8;1 [years; months], range = 6;1–10;11), and compared their L1 language abilities with their age-gender-and grade-matched TD peers acquiring Urdu as L1 and majority language in Pakistan (The majority group; N=31; mean age = 8;1; range = 6;1–10;10). In general, the EM children residing abroad acquire their L1 Urdu under reduced input conditions in their host countries. Whereas children from the majority group living in Pakistan acquire their L1 Urdu under relatively abundant input conditions. It is predicted that reduced input might affect certain linguistic abilities more than other linguistic abilities, and reduced input associated with minority language acquisition context might lead to more restricted age-related progress in the first/heritage language (L1) attrition. Both groups were assessed in a range of outcome measures, encompassing lexical comprehension and production, expressive morphosyntax, narrative comprehension and production, using the Urdu versions of Crosslinguistic Lexical Tasks (CLT, Haman et al., 2015; Hamdani et al., 2020a), and Sentence Repetition task (SRep, Marinis & Armon-Lotem, 2015; Hamdani et al., 2020b), and Multilingual Assessment Instrument for Narratives (MAIN, Gagarina et al., 2019a; Hamdani et al., 2020c), respectively. The results showed that the EM children obtained comparable scores to their majority peers in the domains of narrative comprehension, use of internal state terms, and narrative macrostructure production (story complexity and story structure). On the other hand, the EM children demonstrated significantly poor performance than their majority peers in the lexical, morphosyntactic, and grammaticality (proportion of grammatical C-Units) domains. Although the EM children were similar to their majority peers in terms of exhibiting comparable age-related progress in other language outcome measures, they differed in two narrative microstructure measures indexing syntactic complexity and grammaticality, specifically, mean length of C-units and verb accuracy. In these two measures, the EM children demonstrated restricted age-related progress as compared to their majority counterparts. The overall error analyses also showed that the EM children made significantly more morphosyntactic errors (during narrative production and sentence repetition) and lexical errors than their majority group peers. The findings brace the notion that the effect of language input conditions varies across different language domains. Lexical and morphosyntactic competence are more susceptible to input conditions. Whereas narrative macrostructure is comparatively more resilient to language-specific experiences/knowledge. Study 2 addressed a long-standing issue of differentiating between the effects of language experience and genuine language disorder in multilingual children- a challenging issue in identifying Developmental Language Disorder (DLD) in these children. It presents a case study to demonstrate the potential of the CATALISE diagnostic criteria (Bishop et al., 2017) in combination with the assessment tools from the European Cost Action Language Impairment Testing in Multilingual Settings (LITMUS) battery, to identify language difficulties in a multilingual child with suspected DLD via remote online testing. The participants included one six-year-old (6;8) Urdu-Cantonese EM multilingual child at risk for DLD (S-DLD), and seven age and grade-matched multilingual peers from similar linguistic backgrounds. Their abilities in multiple language areas were assessed via Zoom using Urdu versions of the CLT (Haman et al., 2015; Hamdani et al., 2020a), SRep (Marinis & Armon-Lotem, 2015; Hamdani et al., 2020c), MAIN (Gagarina et al., 2019a; Hamdani et al., 2020c), and CL-NWR (Chiat, 2015; Hamdani et al., 2020d). A nonverbal IQ test (Raven's Progressive Matrices; Raven & Court, 1998) and a parental questionnaire (LITMUS-PABIQ; Tuller, 2015) were also administered. The findings indicated that the child with S-DLD scored significantly lower in multiple language measures compared to her peers in her best/first language, Urdu. Considering also the presence of negative functional impact (language problems affecting her academic and social skills, day-to-day communication) and poor prognostic features (problems persisting beyond age five, problems in multiple linguistic domains, and low receptive skills), and absence of associated biomedical conditions, the results collectively suggested that this participant could be identified as a child with DLD (Bishop et al., 2017). Study 3 also addressed the prevailing issue of differentiating between the effects of language experience and genuine language disorder in multilingual children but focused on Nonword repetition (NWR). NWR has been reported as one of the most sensitive measures in discriminating between children with and without language disorders cross-linguistically. Since nonwords are supposedly unfamiliar words to all children acquiring a certain language, NWR is less influenced by prior linguistic knowledge and therefore could be less biased against multilingual children from diverse language backgrounds who might have less experience of the language targeted during the assessment. This is, therefore, the first empirical study examining the clinical utility of NWR in identifying DLD in Urdu-speaking children. Specifically, it examined whether the new Urdu NWR test would disadvantage TD children with reduced experience in the language of assessment (by comparing TD children with and without reduced experience) and whether this NWR test can also discriminate between children with typical development but reduced experience to the language being assessed and children with DLD. Three groups of children were compared: (i) a group of typically developing (TD) children acquiring Urdu as L1 and majority language in Pakistan (The majority language TD group; N=31; mean age = 8;1, range = 6;1–10;10); (ii) a group of age- gender- and grade-matched TD peers acquiring Urdu as L1 and minority heritage language in Hong Kong who had reduced experience of the language of assessment (The minority language TD group; N=31; mean age = 8;1, range = 6;1–10;11); and (iii) a group of age-matched peers with DLD acquiring Urdu as L1 and majority language in Pakistan who do not have the issue of reduced experience of the language of assessment but suffer from language disorder (The majority language DLD group: N=14; mean age = 8;1, range = 6;1–10;11). NWR performance was assessed by adapting a quasi-universal NWR task from the European COST Action IS0804 LITMUS battery (currently known as the LITMUS Crosslinguistic Nonword Repetition test, CL-NWR) into Urdu (Chiat, 2015; Hamdani et al., 2020d). Results indicated that the majority language DLD group scored significantly lower than both the majority and minority language TD groups at the segment correct level. Moreover, children from the minority language TD group performed similarly to their majority language TD peers suggesting that the new Urdu NWR test did not disadvantage these minority language TD children acquiring Urdu as L1 under reduced input conditions. The findings provided the first evidence that the Urdu CL-NWR test could yield significant TD/DLD group differences, even for TD children with reduced language experience. The three studies in this dissertation are significant in many ways. Findings from Study 1 provide beneficial insights for practitioners and parents working with the EM children related to focused intervention by elucidating their linguistic weaknesses. Findings from Study 2 demonstrated the promise of using the CATALISE framework, the LITMUS battery tasks, and remote online testing in identifying DLD and collecting reference data (i.e., data from a smaller sample to guide developmental expectations when normative data from the larger sample is not available) in multilingual children. Findings from Study 3 demonstrated the diagnostic potential of the Urdu CL-NWR test in identifying DLD in Urdu-speaking children. Urdu, although ranked 10th in the world as a major world language (Eberhard et al., 2022), is lacking in assessment tools. Overall, this dissertation creates four new assessment tools to assess linguistic competence in Urdu-speaking children. Based on their flexibility to be adapted into any language and clinical utility to identify DLD in multilingual children, these new Urdu assessment tools might be practical for many researchers. |
| Rights: | All rights reserved |
| Access: | open access |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/13029