Translation and adaptation of the multilingual assessment instrument for narratives into isiZulu
Date
2024-03
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH SUMMARY: South Africa faces a shortage of linguistically and culturally appropriate speech and language assessments for its multilingual and multicultural population. Clinicians and educators often struggle to differentiate between language impairment, language delay and second-language learning characteristics. The Multilingual Assessment Instrument for Narratives (MAIN) (Gagarina, Klop, Kunnari, Tantele, Valimaa, Balciuniene, Bohnacker & Walters, 2012) was developed for use in multilingual and multicultural contexts, like South Africa, for assessment of narrative comprehension and production. The MAIN’s theoretical basis is the Stein and Glenn (1979) story grammar model, which is hypothesised to be invariant across languages. The goals of this study were to translate and adapt the English version of the MAIN into isiZulu, and to determine the linguistic, cultural and functional equivalence between the two language versions. The study was conducted in three phases. Phase one aimed to translate the English MAIN into isiZulu using a double translation and reconciliation method, guided by the International Test Commission (ITC) Guidelines for Translating and Adapting Tests (second edition). Phase two aimed to describe 5 translation reviewers’ perspectives regarding the linguistic and cultural equivalence of the newly translated Zulu MAIN to the English MAIN. Phase three aimed to determine the functional equivalence of the Zulu MAIN to the English MAIN for the Cat and Dog and Baby Birds and Baby Goats story combinations using the telling elicitation mode. Phase three used a quantitative within-subject, repeated measures study design. Twenty-eight participants in Grade 4 (aged 9-11 years) were recruited from a dual medium school in Katlehong, Gauteng. Inferential statistics was used to analyse the scores for macrostructure, structural complexity, internal state terms and comprehension. The study showed that linguistic and cultural equivalence was achieved by the adaptation of the Zulu MAIN to not only make it equivalent to the English version, but also appropriate for the Zulu population. Translation reviewers pointed out linguistic differences namely, difference in commonality of the Zulu words (linked to varieties of isiZulu), changes in translation and differences in the length of story scripts across the two languages. Phase 3 found that the Zulu and English versions of the MAIN are functionally equivalent for the Cat and Dog story combination only, thus supporting the theory of language invariance of story grammar particularly the Stein & Glenn (1979) model. Participants had higher macrostructure scores, more complex narratives, more internal state terms (ISTs) and higher comprehension scores for the Baby Birds and Baby Goats narratives that they told in isiZulu than those they told in English. The study found that the Baby Birds and Baby Goats for isiZulu and English are not functionally equivalent and recommended the use of the Cat and Dog story combination for bilingual comparison.
AFRIKAANSE OPSOMMING: Geen opsomming beskikbaar.
AFRIKAANSE OPSOMMING: Geen opsomming beskikbaar.
Description
Thesis (MSL&HT)--Stellenbosch University, 2024.