In line with relevance and complexity of language and reading assessment in bilingual children, this work incorporates recommended language assessment of bilingual children’s first as well as their second language and making use of innovative technologies. Previous research relates to the suitability of tasks assessing language with language-specific clinical markers, and reading skills in the bilingual population across languages. Some of these tasks have been implemented as computerized tasks allowing for automatic administration and evaluation. However, these computerized tasks are restricted to the characteristics defined by developers and cannot be adapted for use in school or clinical settings. This manuscript includes a review of the scientific literature concerning language and reading screening approaches and six empirical studies on the construction, administration, suitability and useful-ness of computerized bilingual language and reading screenings. The first study concerns how Speech and Language Therapists assess bilingual children. In a survey with 300 Speech and Language Therapists it was investigated how far Speech and Language Therapists consider assessment of both languages spoken relevant and realizable with computerized screening tasks. Results suggest an imbalance between the knowledge of the specific requirements for assessment of multilingual children and their application at work. Speech and Language Therapists also indicated open-ness towards computerized solutions. The second study investigates the potential of reading assessment in both languages through automatic task administration and evaluation. The performance in standardized Italian reading tests of 33 Mandarin-speaking children living in Italy, attending grades 3 and 4 in primary school, is compared to task performance in computerized screening tasks, implemented on E-Prime 3.0, administered in presence of the examiner. Results suggest that computerized reading and language tests reliably assess reading performance in L2 speakers of Italian. The third, fourth, fifth and sixth study investigate the potential of the modifiable screening platform MuLiMi, created for this dissertation, in the risk identification of Developmental Language Disorders and Developmental Dyslexia risk in bilingual children. These studies covering five language groups, two age ranges and two different screening purposes, testing 109 child participants indicate that it is possible to automatically assess language and reading skills in both languages with the screenings implemented on the MuLiMi web app. Further interesting findings suggest an advantage of paradigms such as “Who says it right?” over judgement tasks for screening tasks on morphosyntactic processing and phonological awareness as well as to the potential difference in contribution to risk identification between different item subcategories, e.g. verbs vs. noun comprehension and language-specific vs. non-language-specific nonwords. In addition to that, study six assesses the usability of this screening platform as perceived by the examiners administrating the screenings and by the children. The results generally suggest user-friendliness and ease in use even when administering these screenings remotely. Some of the information obtained indicate the need of improvement concerning the system speed. In summary, findings concern the possibility of using fully automatically administered and scored tasks for the identification of bilingual children at risk for language or reading disorders, meeting the requirements concerning reliability and validity of the tests and allowing an assessment perceived as easy, useful and pleasant. The application of the screening platform MuLiMi appears to meet the requirements concerning bilingual language and reading assessment, allowing for the comparison of both languages as well as conscious decision-making for diagnoses and intervention need.

Considerando la rilevanza e complessità della valutazione del linguaggio e della lettura nei bambini bilingui, il presente lavoro raccomanda di valutare il linguaggio nella prima e nella seconda lingua e l’utilizzo di tecnologie innovative. Ricerche precedenti hanno indagato, in varie lingue, l'adeguatezza di prove che valutano il linguaggio utilizzando marcatori clinici lingua-specifici, e la lettura nella popolazione bilingue. Alcune di queste prove sono state implementate in forma computerizzata per la somministrazione e la valutazione automatizzate. Spesso queste prove computerizzati di screening sono limitate alle caratteristiche definite dagli sviluppatori e non possono essere adattate per un utilizzo in scuola o clinica. Il presente elaborato include una revisione della letteratura scientifica relativa agli strumenti di screening del linguaggio e della lettura e sei studi empirici riguardanti la costruzione, la somministrazione, la validità e l'utilità di screening computerizzati di linguaggio e lettura nei bambini bilingui. Il primo studio riguarda i metodi con cui i logopedisti valutano i bambini bilingui. Attraverso un questionario compilato da 300 logopedisti è stato indagato se loro considerano la valutazione di tutte le lingue parlate rilevante e realizzabile attraverso dei compiti computerizzati. I risultati suggeriscono uno squilibrio tra conoscenze delle prassi raccomandate e la loro effettiva applicazione. I logopedisti si dichiarano aperti alla possibilità di utilizzare delle procedure computerizzate. Il secondo studio indaga il potenziale della valutazione della lettura in entrambe le lingue attraverso procedure computerizzate. Confrontando le prestazioni di 33 bambini, parlanti mandarino-italiano e frequentando la terza e quarta elementare in Italia, in test standardizzati di lettura in italiano con quelle conseguite in compiti di screening computerizzati, implementati con E-Prime 3.0, somministrati in presenza, emerge che i test computerizzati valutano in modo affidabile le competenze di lettura nei parlanti di italiano L2. Il terzo, quarto, quinto e sesto studio indagano il potenziale della piattaforma di screening modificabile MuLiMi, creata per questo lavoro di tesi, nell'identificare il rischio di Disturbo Primario del Linguaggio e di Dislessia Evolutiva in bambini bilingui. Questi studi includono cinque gruppi linguistici, due fasce di età, due target di studio e 109 bambini e confermano che è possibile valutare le competenze di linguaggio e lettura in entrambe le lingue attraverso la procedura di screening implementata sulla web app MuLiMi. Altri risultati suggeriscono un vantaggio di paradigmi come “Chi lo dice giusto?” rispetto ai giudizi di correttezza nelle prove morfosintattiche e di consapevolezza fonologica, e la potenziale differenza nel contribuire all'identificazione del rischio da parte di diverse sottocategorie di item, per esempio comprensione di verbi vs. nomi e nonparole lingua-specifiche vs. non lingua-specifiche. Il sesto studio indaga l'usabilità di MuLiMi analizzando le impressioni degli esaminatori e dei bambini valutati. I risultati suggeriscono che entrambi i gruppi trovano l’uso della piattaforma facile, anche utilizzata a distanza. Alcune delle risposte indicano la necessità di migliorare la velocità del sistema. In conclusione, i risultati riguardano la possibilità di utilizzare compiti somministrati e corretti in modo automatico per l'individuazione dei soggetti bilingui a rischio per disturbi di linguaggio e di lettura, rispettando i requisiti di affidabilità e validità delle prove e consentendo una valutazione valutata come facile, utile e piacevole. L'utilizzo di MuLiMi sembra soddisfare i requisiti relativi alla valutazione del linguaggio e della lettura nei bilingui, consentendo il confronto tra le competenze in entrambe le lingue, e favorendo negli esaminatori un processo decisionale consapevole per la diagnosi e l’intervento.

(2022). Computerized system for bilingual language and reading screenings. (Tesi di dottorato, Università degli Studi di Milano-Bicocca, 2022).

Computerized system for bilingual language and reading screenings

EIKERLING, MAREN REBECCA
2022

Abstract

In line with relevance and complexity of language and reading assessment in bilingual children, this work incorporates recommended language assessment of bilingual children’s first as well as their second language and making use of innovative technologies. Previous research relates to the suitability of tasks assessing language with language-specific clinical markers, and reading skills in the bilingual population across languages. Some of these tasks have been implemented as computerized tasks allowing for automatic administration and evaluation. However, these computerized tasks are restricted to the characteristics defined by developers and cannot be adapted for use in school or clinical settings. This manuscript includes a review of the scientific literature concerning language and reading screening approaches and six empirical studies on the construction, administration, suitability and useful-ness of computerized bilingual language and reading screenings. The first study concerns how Speech and Language Therapists assess bilingual children. In a survey with 300 Speech and Language Therapists it was investigated how far Speech and Language Therapists consider assessment of both languages spoken relevant and realizable with computerized screening tasks. Results suggest an imbalance between the knowledge of the specific requirements for assessment of multilingual children and their application at work. Speech and Language Therapists also indicated open-ness towards computerized solutions. The second study investigates the potential of reading assessment in both languages through automatic task administration and evaluation. The performance in standardized Italian reading tests of 33 Mandarin-speaking children living in Italy, attending grades 3 and 4 in primary school, is compared to task performance in computerized screening tasks, implemented on E-Prime 3.0, administered in presence of the examiner. Results suggest that computerized reading and language tests reliably assess reading performance in L2 speakers of Italian. The third, fourth, fifth and sixth study investigate the potential of the modifiable screening platform MuLiMi, created for this dissertation, in the risk identification of Developmental Language Disorders and Developmental Dyslexia risk in bilingual children. These studies covering five language groups, two age ranges and two different screening purposes, testing 109 child participants indicate that it is possible to automatically assess language and reading skills in both languages with the screenings implemented on the MuLiMi web app. Further interesting findings suggest an advantage of paradigms such as “Who says it right?” over judgement tasks for screening tasks on morphosyntactic processing and phonological awareness as well as to the potential difference in contribution to risk identification between different item subcategories, e.g. verbs vs. noun comprehension and language-specific vs. non-language-specific nonwords. In addition to that, study six assesses the usability of this screening platform as perceived by the examiners administrating the screenings and by the children. The results generally suggest user-friendliness and ease in use even when administering these screenings remotely. Some of the information obtained indicate the need of improvement concerning the system speed. In summary, findings concern the possibility of using fully automatically administered and scored tasks for the identification of bilingual children at risk for language or reading disorders, meeting the requirements concerning reliability and validity of the tests and allowing an assessment perceived as easy, useful and pleasant. The application of the screening platform MuLiMi appears to meet the requirements concerning bilingual language and reading assessment, allowing for the comparison of both languages as well as conscious decision-making for diagnoses and intervention need.
GUASTI, MARIA TERESA
LORUSSO, MARIA LUISA
DPL; Dislessia Evolutiva; bi-/multilinguismo; valutazione; tecnologia
DLD; Dev. Dyslexia (DD); bi-/multilinguism; assessment; tecnologia
M-PSI/04 - PSICOLOGIA DELLO SVILUPPO E PSICOLOGIA DELL'EDUCAZIONE
English
15-lug-2022
PSICOLOGIA, LINGUISTICA E NEUROSCIENZE COGNITIVE
34
2020/2021
embargoed_20250715
(2022). Computerized system for bilingual language and reading screenings. (Tesi di dottorato, Università degli Studi di Milano-Bicocca, 2022).
File in questo prodotto:
File Dimensione Formato  
phd_unimib_849495.pdf

embargo fino al 15/07/2025

Descrizione: Tesi di Eikerling Maren 849495
Tipologia di allegato: Doctoral thesis
Dimensione 4.37 MB
Formato Adobe PDF
4.37 MB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/392352
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
Social impact